Jllllll Illlllllllllllllllflllllllllllfl'llll L, L

3 1293 00104 9893

:.,
LIBRARY
Michigan State

University

 

 

 

This is to certify that the
dissertation entitled
CAUSAL MODELING OF PARENTING ATTITUDES IN
MOTHERS WITH SEVERE PSYCHOPATHOLOGY

presented by

Fred Arthur Rogosch

has been accepted towards fulfillment
of the requirements for

Ph .D . degree in Jim

 

/.

 

Mag» Jofessor

Date 9" 14-87

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

MSU

LIBRARIES
m

\-

 

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

APWBD'Q‘??‘

120

 

 

CAUSAL MODELING OF PARENTING ATTITUDES IN

MOTHERS WITH SEVERE PSYCHOPATHOLOGY

BY

Fred Arthur Rogosch

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Psychology

1987

’I1

"I...

 

0"...
“.4;
'2‘.:
‘1'.
.
I'a .‘I
.
"u \
'i
t
I “‘
.0
I
H
a
:-
.'
I.~ :
U
‘ \
I
D

 

ABSTRACT

CAUSAL MODELING OF PARENTING ATTITUDES IN
MOTHERS WITH SEVERE PSYCHOPATHOLOGY

BY

Fred Arthur Rogosch

Parenting by mothers with severe mental disorders
constitutes a critical challenge to these mothers'
ongoing psychological adjustment and to the develOpment
of their children. The extent of potential risk for
parenting difficulty is likely to be influenced by the
mother's develOpmental history, personality, social
context, and her child's behavioral adjustment. In the
present research, a causal model of parenting in normal
mothers was expanded in order to investigate determi-
nants of individual differences in parenting attitudes
and styles within a group of mothers with schizophrenic
or major affective disorders.

The participants Lfl=48), hospitalized in one of
four state psychiatric facilities, were mothers who had
at least one child age 12 or younger with whom they had
lived prior to admission. Data were collected through
interviews conducted by trained graduate students.

The results indicated that child behavioral

 

 

..AI.

 

 

 

 

 

   

 

 

 

 

 

 

33-”

‘ i
.“

Ill

9 t

.

I‘ ::
‘~

 

adjustment and living situation adequacy did not predict
parenting style. Path analysis demonstrated that fre-
quent relapse and lax discipline or uninvolvement from
one's own mother provided the strongest direct influ-
ences on parenting style and contributed to two major
causal pathways. In the first path, frequent relapse
constituted a direct influence which was relatively
distinct from other influences. This suggests that
greater severity and chronicity of disorder, with the
resulting frequent separations from one's children, have
a detrimental effect on parenting. In the second path,
childhood relational experiences (absence of a close
relationship, separation from one's own mother, and
maternal lax discipline or uninvolvement) directly
influenced self esteem and emotional support, and these
latter factors directly influenced parenting style. Lax
discipline also provided a strong direct influence. The
second causal path was interpreted in an attachment
framework. Internal representational models of
attachment relationships, developing in the mother's own
childhood, were conceptualized as contributing to
responsive or unresponsive parenting in the current
relationship with her own children and to the level of
self esteem and involvement with emotionally supportive
others. Research limitations, future directions, and

implications for treatment were discussed.

ACKNOWLEDGMENTS

I would like to express my deep appreciation to Dr.
Anne Bogat for her consistent support and guidance
through all of the phases of this dissertation. Dr.
Bogat was always a source of encouragement and direction
in sorting through the numerous conceptual and
logistical dilemmas associated with this research. She
facilitated bringing clarity and focus to my ramblings
and frequently served as a voice of reason. Dr. Bogat:
through her encouragement and honesty, is both mentor
and friend.

This research would not have been possible without
the support of Dr. Carol Mowbray and the Michigan
Department of Mental Health. I am indebted to Dr.
Mowbray for her constant patience and tolerance for the
delays in bringing this research to completion. Dr.
Mowbray has been an inspiration through her commitment
and dedication to addressing the needs of chronically
mentally ill women, and she has provided me with an
excellent model of how to use research to generate
necessary social change.

I wish to thank Dr. Robert Zucker for his

iv

, Q

rm.“
‘

 

.4:

71'
.Io

 

.\~
ul-

n... .

.

'-

Ms

 
 

 

thoughtful recommendations and suggestions for improving
this research. I appreciate Dr. Zucker's high standards
and balanced evaluations which promote soundly grounded
work. I am grateful for Dr. Zucker's clinical
supervision which allowed me to learn much about the
experience of family dynamics.

Throughout my graduate training, Dr. Gary Stollak
has provided alternate perspectives on a variety of
tepics, and he has encouraged critical questioning of
just what psychologists do. Dr. Stollak's expertise
with issues related to parenting helped me to formulate
the central parenting features examined in this
research.

I am indebted to Karen Williams, Ann Wagner, Carol
Schwartz, Wendy Sabbath, Ralph Tobias, and Sue Popovich,
fellow graduate students who generously volunteered
their talents for interviewing the mothers. Their
enthusiasm and the perceptions and emotions we shared
made the arduous interviewing experience more
meaningful. I also appreciate Sue Popovich's assistance
in the coding of the parenting measure.

The friendly cooperation of the hospital staff
members at each of the participating psychiatric
facilities made the data collection proceed as smoothly
as possible. Special thanks are owed to Jerry Leismer,

Nelia Deiz, Ken Janssen, Ralph Blair, and Robert Yesner

‘A'
.n

F!!!
n.

q
I
O

 

 

 

for their support and their assistance in coordinating
the contacts with the patients.

Finally, I wish to thank the mothers who
participated in this research. I respect them for their
courage and willingness to share personal aspects of
their troubled lives, while receiving very little in

return.

vi

TABLE OF CONTENTS

Page

List of Tables . . . . . . . ix
List of Figures . . . . . . . xi
List of Appendices . . . . . . . xii
INTRODUCTION . . . . . . . . 1
Developmental History and Personality . . 8
Maternal Developmental History . . . 8
Maternal Personality . . . . . 14
Social Contextual Influences . . . . 16
Social Networks . . . . . . 16
Physical Living Environment . . . . 21
Child Characteristics. . . . . . 22
Research Goals . . . . . . . 22
METHOD . . . . . . . . . 30
Subjects . . . . . . . . 30
Materials . . . . . . . . 32
Current Life Situation Questionnaire . . 32

Early Life Questionnaire . . . . 33
Child Reports of Parental Behaviors Inventory 34
Symptom Checklist - 10 . . . . . 35

Self Esteem Scale . . . . . . 36
Social Support Questionnaire . . . 36
Mother-Child Relationship Evaluation . . 38
Parent Attitude Survey Scales . . . 4O
Sensitivity to Children Questionnaire . . 41
Child Behavior Checklist . . . . 45
Observer Checklist . . . . . 45

Case Record Review . . . . . 46
Procedure . . . . . . . . 47

vii

RESULTS 0 O C D 0

Experience in Family of Origin

Influences on the Onset of Parenting

Current Parenting Context .
Social Support . .

Psychological Distress and Self Esteem
Perception of Child Behavior Problems

Measures of Parenting .
Path Modeling of STCSTYLE .

DISCUSSION . . . .

Evaluation of the Hypothesized
Proximal Models of Parenting
Distal Models of Parenting
Determinants of Parenting:
An Attachment Perspective
Attachment in Childhood and
Other Parenting Measures
Education and Parenting .
Physical Punishment . .
Potential Sources of Stress
and Support in Parenting

50

5o
52
54
57
60
64
64

76
96

Model of Parenting 96

Research Limitations and Future Directions

Treatment Implications .
APPENDICES . . . .
LIST OF REFERENCES . . .

viii

100
103

108
111
1fl+
115
116
121
124
128

174

10.

ll.

12.

13.

14.

LIST OF TABLES

Multiple Regression to Predict
Age at First Birth . . . . .
Frequencies of Marital Status by Race .

Intercorrelations of Types of Social Support
Multiple Regression to Predict Self Esteem

Multiple Regression to Predict
Psychological Distress (SCL-lO) . .

Multiple Regression to Predict
Perception of Behavioral Difficulty
in the Child O O O O O O

Intercorrelations of Parenting Measures .

Multiple Regression to Predict
Overprotection-Overcontrol (OVPOVC) .

Multiple Regression to Predict
Hostile-Objectification (HOSOBJ) . .

Multiple Regression to Predict
Parenting Difficulty . . . .

Multiple Regression to Predict
Physical Punishment . . . .

Potential Variables for Path Models
in Each Domain . . . . .

Intercorrelations of Variables
in the Path Models . . . . .

Decomposition of Relationships between
Independent Variables and Parenting Style

ix

Page

53
55

62

63

65

68

7O

72

73

75

77

81

86

 

Table Page

15. Decomposition of Relationships of Predictor
Variables with STCSTYLE into Direct
Effects, Indirect Effects, and Spurious
and Unanalyzable Effects for the
Path Models A2, 82, and C2 . .

 

 

Figure

1.

LIST OF FIGURES

Belsky's (1984) Conceptual Model of
the Determimnts of Parenting . . .

Path Model of the Determinants of
Parenting in Severely Disordered Mothers

Conceptual Organization of Parenting

Styles . . . . . . .
Path Models of Proximal Variables . .
Path Model A2 . . . . . .
Path Model 32 . . . . . .
Path Model C2 . . . . . .

Page

25

43
83
88
89
9o

 

Appendix

LIST OF APPENDICES

Current Life Situation Questionnaire
Early Life Questionnaire

Child Reports of Parental Behavior

Inventory .

Symptom Checklist - lO .

Self Esteem Scale

Social Support Questionnaire
Mother-Child Relationship Evaluation
Parent Attitude Survey Scales
Sensitivity to Children Questionnaire

Child Behavior Checklist

Observer Checklist

Consent Forms

xii

Page
128

134

137
140
141
142
146
149
150
155
163
168

INTRODUCTION

How mothers with severe mental disorders function
as parents will have a profound impact on the
development of their children. While this has been
generally acknowledged, inadequate attention has been
given to the influences that parenthood may have on the
adjustment of the disordered mothers themselves and to
the processes that affect the degree of assumed
impairment in their parental functioning. Test and
Berlin (1981) have noted that the special needs of
chronically mentally ill women who are mothers have not
been recognized, and consequently, the provision of
services to assist these women in their parenting role
has been largely neglected. This general failure to
provide support to these women as mothers suggests that
they will be at continued risk to function with
difficulty in this major life role. As a result,
successful adjustment to community living may be
undermined, contributing to cyclic hospital admissions.
Not only will parenting difficulties exacerbate the
adjustment of disordered mothers, but it also will
detrimentally influence the development of the mothers'
children, potentiating mental health problems in the

future. In order to begin to address these concerns,

 

the present investigation sought to examine in detail
potential processes contributing to variations in
parenting attitudes and effectiveness among severely
disordered mothers of minor children.

Previous investigations of families containing a
member with severe psychopathology (i.e., schizophrenic
disorder or major affective disorder) have tended to
follow one of two approaches, neither of which has
specifically focused on the mothers with mental
disorders and their parenting. First, extensive
research programs have studied the children of mentally
ill parents (e.g., Watt, Anthony, Wynne, & Rolf, 1984;
Beardslee, Bemporad, Keller, & Klerman, 1983: Sameroff,
Seifer, & Zax, 1982: Garmezy, 1974). The major concern
in this ”high risk” research paradigm has been the
investigation of the increased potential of these
children to develop their own psychological disorders as
a result of genetic heritage and, secondarily, familial
influences. Research relevant to parenting has compared
parent characteristics and practices of the risk groups
to that of the general population. Little attention has
been given to the potential within-group variation in
the parenting provided by the severely disordered
parents or to the determinants of this parenting, both
of which may contribute to variations in the degree of

vulnerability in the children.

The second main area of research has investigated
the interaction patterns and communication in families
in which an individual family member has a severe mental
disorder (e.g., Mishler & Waxler, 1983: Wynne, 1981:
Leff & Vaughn, 1985). Typically, the child, usually an
adolescent or young adult rather than a parent, has been
the family member with the disorder, and a broad range
of family processes thought to contribute to
symptomatology in the target offspring has been
investigated.

Thus, in these two bodies of family research on
psychopathology, the attention has focused on the
children of disordered parents or the families of
disordered children. The present research with its
focus on parenting in disordered mothers falls at a
midpoint between these generational levels. The
mother's disorder and her functioning as a parent may,
in part, stem from her experiences in her family of
origin. Thus, the high risk and the family interaction
research areas have important implications for the
parenting of severely disordered women. However, the
investigation of specific determinants of parenting
within this group of mothers has been relatively
neglected aside from the work of Cohler and Grunebaum
and their colleagues (e.g., Grunebaum, Weiss, Cohler,

Hartman, & Gallant, 1982).

 

 

‘ywan
.

I. g..

'I
I'l-

nou u

'3‘;
311

n ,
u .!

 

" o.

.2’.

I
we...

 

 

'5
I.‘

a...-

 

 

 

 

q
n

q
H

‘y

In comparing mentally disordered mothers to
nondisturbed mothers, differences in parenting attitudes
and behavior have been observed (Cohler, Gallant,
Grunebaum, Weiss, & Gamer, 1980). Disordered mothers
did not value reciprocal communication with the child,
had difficulty distinguishing their own needs from those
of the child, and tended to deny concerns about child
care. Further, these mothers exhibited apathy and
restricted affect while interacting with their children.
Klehr, Cohler, and Musick (1983) have discussed
difficulties in parenting emanating from maternal
psychosis including problems in establishing a healthy
relationship with a child, involvement of the child in
the mother's pathological process, using the child as a
target of the mother's impulses, and creating confusion
for the child by vacillation in the mother's attitudes
and mood. Weissman and Paykel (1974) have noted
parenting difficulties in mothers with depression: two
patterns were prominent in these mothers' interactions
with their children: hostility and irritability vs.
overconcern, helplessness, and guilt. Collectively,
these descriptions highlight some of the problems in
parenting experienced by disorderd mothers as contrasted
with nondisturbed mothers. However, these group
contrasts do not reveal the range of differences within

the group of disordered mothers: not all disordered

""ers e

...> 1
ll..\."|.
noeaUUO'

- A
an ‘1
5.. camel
.._: in.&\

m";
rat-Iran

31532356

1 .
' I..‘A q
“.“VI‘VI 4

'4'.

l

l .
‘01:

fezm

I,
“'5."
i . A .
‘I“.Vn
‘

L.“ {1‘
' l4, 1'
.Je'

mothers evidence the same degree of parenting
difficulty. For instance, Kauffman, Grunebaum, Cohler,
and Gamer (1979) reported that high social competence in
the children of disordered mothers was found among those
children who had warm relationships with their mothers
and whose mothers evidenced more adjusted social
functioning.

The study of parenting determinants in nondisturbed
mothers provides conceptual direction for inquiry into
the processes contributing to individual differences in
parenting among disordered mothers. Belsky (Belsky,
1984: Belsky, Robins, & Gamble, 1984) organizes a
multiplicity of factors related to parenting in normal
parents into a useful conceptual model of the
determinants of individual differences in parental
functioning (see Figure l). The basic components of his
model include three main factors listed in order of
importance: (1) parental personality and psychological
well-being which are, in part, determined by
developmental history: (2) contextual subsystems of
support (marital relations, social network, employment):
and (3) individual child characteristics. In this
model, parental personality has both a direct influence
on parenting and an indirect influence through its
relationship to marital adjustment, the social network,

and employment. These latter variables, in turn, have

C I

f

.\~u.n' \

II‘ J
In!
I I. A

~D.§-¥‘- CV

 

acuemoao>oa
uaaco

 

hnovma:
H.a:osmoao>on

       

hpaamcomwom

     

 

    

mcaacmnsm

 

 

 

no amauoaoaumno

 
  

  

mnoaumaom
Hawahmz

 

 

xuozaoz
Hdaoom

 

 

 

 

 

Esteem no mBSAEBS on... no 88: $368.86 $33 6.338
.s enema

11:91:: I
erlcyzent
gerscnalit
granting.
5x: as te
garents at
all exni:
axializa:

.:;::ents

-.
. V
A KAPQ

I.‘

W“ H
“‘39

ii.

‘u=:§

their own direct influences on parenting. Further, the
contextual variables (marital relations, social network,
employment) also reciprocally influence parental
personality and thereby also have indirect influences on
parenting. Finally, individual child characterstics
such as temperament place different demands on the
parents and thus the type of parenting behaviors they
will exhibit. Belsky noted that it is rare in the
socialization literature for multiple links between
components to be investigated in the same study. As a
result, the complex intervening processes contributing
to parenting are seldom examined.

Although Belsky developed his model to examine
differences in parenting among normal as well as
child-abusing parents, the basic components of his model
can be used to study the multiple determinants of
variations in parenting among mothers with severe mental
disorders. Although normative patterns of influence
form the basic framework, they need to be elaborated or
supplemented to incorporate influences more likely to be
unique to disordered mothers (i.e., effect of
hospitalization). The range of influences to be
considered include the following: develOpmental history,
including a history of separation and loss and
smrceptions of maternal parenting received as a child:

personality adjustment, including severity and

.‘mn M. '
,iJV-JO". ‘

self-este

SCCLCECC.’

if” nus.‘
tub Mil.

..
I-'
3.351le

.5’5'1.‘ ,
f-‘uuol \
.

2121.15 1

::ese dil

' a
-

.;:fe:en(

\

chronicity of disorder, type of disorder, and
self-esteem: social support networks, including
relationships with her family of origin, extent of
emotional support, support in child rearing:
socioeconomic status and physical living environment:
and individual child characteristics. In addition to
establishing how these features are related to
parenting, interrelationships between the separate
domains must also be explored in order to understand how
these diverse influences collectively promote individual
differences in parenting.
Developmental History and Personality
Maternal Developmental History. Several factors in
the mother's own development are important for both her
parenting as well as her psychological well-being.
Notable among these factors is a history of separation
from or loss of important attachment figures in her
childhood, a perspective derived from attachment theory
(Bowlby, 1969: 1973: 1980). Frommer and O'Shea (1973a:
.1973b) have shown that women who experienced separation
from their parents when they were children had more
difficulty mananging their own infants as compared to
women without such separations. Further, many of these
nurtlaers also evidenced depression and had more severe
difficulty in managing their children if current marital

r913 tions and housing conditions were poor. SimilarIYI

.. .. .1 .f.‘
l:..(6.wl

5.251 c'r

522131 c
mail
fine”;

II).F- fl
:vbab mie

‘|,..
L.'= Dee
"i. s

l’:
"‘5 a
19..
n‘“
.
, . up
\a‘en g
‘I

Wolkind, Hall, and Pawlby (1977) observed mothers, who
had a childhood history of separation and loss,
interacting with their young infants. These researchers
found that the mothers engaged in less physical and
social contact with their infants, suggesting that
aberrations in the current mother-infant attachment were
developing. Thus, childhood disruptions in the
attachment relationships experienced by mothers may have
an influence on the later attachment relationships that
develop with their own children. Wolkind et al.'s work
also emphasized that the mother's disrupted attachments
in childhood resulted in the current problematic
attachment relationships because of a more pervasive
disruption in the family of origin as a result of
separation or loss rather than the separation or loss in
and of itself.

While disruption of the attachment relationship has
thus been shown to influence parenting with one's own
children, the quality of the attachment relationship or,

‘more generally, the parent-child relationship, is also
influential. Evidence for this relationship can be
found in primate studies of adult monkeys raised in
deviant or isolated early environments (Harlow and
Barlow, 1969). As adults these monkeys exhibited
grossly abnormal mothering behavior: some failed to

nurse, avoided contact, or rejected their babies. Some

:f these

‘:neir

c n
v

:f the ef
mes fro
fzziing i
fre;;en:l

garezts (

10

of these primate mothers were also violent and abusive
of their offspring. Among humans, the clearest example
of the effect of disturbed parent-child relationships
comes from the child abuse literature. A consistent
finding is that parents who abuse their children very
frequently report an abusive relationship with their own
parents (Belsky, 1980: Rutter, 1981). Undoubtedly,
one's parents have provided powerful models of how to
parent, and much of this learning would be expected to
be utilized with one's own children. Further, more
appropriate patterns of parenting would be less likely
to have been learned if a mother did not experience such
patterns in her own childhood.

Rutter and Madge (1976) discuss the cycle of
intergenerational transmission of styles of parenting
and conclude that continuity in parenting across
generations is more likely with deviant, severe patterns
of parenting than it is with nondisturbed or normal
patterns. Nondeviant parenting is more likely to be
influenced by the current context and the values of the

tinues, resulting in less intergenerational continuity in
specific nondisturbed parenting practices. For mothers
with severe psychopathology, the likelihood of there
being continuity in parenting with the parenting they
experienced would be greatest if their experiences with

thelr parents were severe and aversive. Recent research

.'AA
l..vV

';v;n

to. ...

P-q o
.1...

 

 

run

it.

11

(Crook, Raskin, & Eliot, 1981) has demonstrated that
hospitalized depressed women report considerable
maternal rejection in their own childhoods. The
literature on the relationship between poor parent-child
relations and psychopathology is vast and is not the
main interest here. Rather, if an increased likelihood
that women with severe disorders experienced deviant
parenting in their childhood can be assumed, it follows
that these mothers might be more likely to repeat such
parenting with their own children. For example, Hilgard
and Fisk (1960) describe patterns in which daughters of
mentally ill parents recapitulate the parenting
experiences they had as children with their own
children.

Benedek has been influential in theorizing about
how the early experience of being parented influences
adult personality develOpment when an individual becomes
a parent (Benedek, 1959: Parens, 1975). The presence of
a child often results in the resurrection of issues from

the mother's own childhood. Benedek maintains that this
ma)? provide an opportunity for reworking earlier
experiences and lead to resolutions of conflictual
issues and personal growth. On the other hand, more
ovceIWJhelming and traumatic experiences may also be
rein troduced into the mother's awareness. Not only will

“11:3 be psychologically stressful, but she may also fear

6-30
‘... b

ii'jS

':",(‘L‘
'1' in
- a

5791's

' 5-.

(.114

Bit

" bu

 

 

A
.,I
‘u

56

.1.‘
1..

 

 

12

that she will respond to her child in the inappropriate
ways she may have experienced.

However, repetition of negative parenting
experiences does occur. Several authors have suggested
psychological processes to account for continuity in
aversive parenting. Fraiberg, Adelson, and Shapiro
(1975) discuss the aversive childhood experiences of
mothers who exhibit unresponsive and rejecting parenting
of their own infants. They report that these mothers
reenact disturbed relationships with their own infants
consistent with the experiences they had with their own
mothers. The critical element predisposing these
mothers to repetition was repression of the affective
experience of their relationships with their mothers.
Similarly, Main and Goldwyn (1984) have demonstrated
that mothers, whose own mothers were rejecting, were
also rejecting of their infants, and current
difficulties in mother-infant attachment were evident.
These mothers tended to distort, disorganize, and
exclude from awareness information about their

chiJldhoods. Their current perspective of their mothers
was idealized in a defensive fashion. In both of these
reports, mothers exhibited difficulty in accurately
representing their own childhood relationships with
their mothers, and as a result of repression and

distortion were prone to repeat their own eXperiences.

13

Boszormenyi-Nagy and Spark (1973) also describe how
parents may repeat the parenting patterns they
experienced as a means of maintaining loyalty to their
parents. By restructuring negative parenting
experiences as appropriate and utilizing them with their
current children, parents may protect the image of their
parents as caring.

In addition to the effects on parenting, disrupted
attachment in childhood has also been related to
depression (Bowlby, 1969: 1973: 1980). This view has
received support from Brown and his colleagues (Brown,
Harris, & Bifulco, 1986: Brown, 1982: Brown & Harris,
1978: Brown, Bhrolchain, & Harris, 1975). Among female
psychiatric patients, 22.4% evidenced loss of their
mother before age 10 whereas only 6% of the control
sample evidenced such a loss. This relationship was
even more pronounced among depressed women of the lower
social class. Maternal loss was likely to have resulted
in subsequent deficiencies in care received as a child,
iJuflicating that the loss initiated a number of

detrimental influences compromising development and
contributing to later maladjustment. In addition to
maternal loss, Brown (Brown & Harris, 1978: Brown,
Harris, & Bifulco, 1986) has shown three other factors
to '3‘? important: lack of employment, low intimacy with a

spouse or boyfriend, and having three or more children

14

under 14 years of age. These researchers found that a
multiplicative relationship among these factors rather
than a linear, additive effect was operative. The fact
that having multiple young children is a risk factor for
depression suggests that parenthood itself may become a
risk factor for depression. In fact, among a group of
female twins, the prevalence of psychiatric disorder was
greater among those women with children than among their
childless twin sisters (Malmquist & and Kaij, 1960).
In summary, the research of Brown and his colleagues has
explored how maternal loss and separation contribute to
the interplay of a number of factors to potentiate
depression, rather than examining loss as a direct
influence in isolation from its influence on other
features of the mother's life. Thus, the causal
relationships between independent variables as well as
their direct influence on depression are examined. The
work of Brown and his colleagues on depression in women
has not focused on parenting itself or more general
.Lssues of psychOpathology in which depression may
exist.

Maternal Personality. Among mothers with severe
psychological disorders, more research has been
conducted in the high risk tradition with schizophrenic
mothers and their children than with mothers with

affeecrtive disorder. Yet, as shown by Sameroff, Seifer,

15

and Zax (1982), the severity and chronicity of the
mother's disorder is more strongly related to child
problems, and by inference to parenting, than is the

mother's diagnostic group. Contradictory findings by

Cohler and Musick (1983) indicate that parenting appears

more deviant in psychotic depressed mothers than in
schizOphrenic mothers. Further, children of
schizophrenic mothers were less distinguishable from
normal controls than were children of the depressed
mothers. Grunebaum et a1. (1982) note that the
chronicity of the mother's disorder, as indicated by
multiple psychiatric hospitalizations, results in
multiple separations between the mother and her
children. Upon reunion, both mother and child have
difficulty in reuniting and the mother's guilt over
being away from her children may make her particularly
anxious in her child care abilities.

Another important individual feature of the
mother's personality is her level of self esteem which
may'act as a personal resource to her in dealing with

stressful life situations. Feelings of self worth and
self acceptance contribute to a mother's ability to be
responsive toward and accepting of her children. Ricks
(1985) provides evidence that mothers of securely
atJT-a'lthed infants had higher self esteem and more

P°sitive recollections of childhood relationships than

 

it 13:31

arris (1
actsrs (
nth a 3;
may fcur

1' " '
.1331: 3nd

...';~ H
““"a 5U:

:E“ ‘\
"-56.4; h

16

did mothers of anxiously attached infants. Brown and
Harris (1978) theorize extensively about the importance
of self esteem in relation to the four vulnerability
factors (maternal loss, unemployment, lack of intimacy
with a spouse, and multiple children under age 14) which
they found to be predictive of depression.
Additionally, Rosenberg (1965) found that individuals
with low self esteem were likely to report feeling that
their parents were uninvolved with them as children.

Thus, the complications of mental disorder and
indicators of general personality adjustment (i.e., self
esteem) have been linked to parenting. However, social
contextual features have not been frequently
incorporated into the associations between
psychopathology or personality and parenting while also
considering historical developmental influences.
Social ContextugliInfluences

Social Support Networks. A mother's social support
network may contribute not only to her psychological
adjustment by providing sources of validation and
acceptance, but also may buffer or enhance her
parenting. Leavy (1983) has reviewed the relationship
of social support to psychological adjustment and
reported that clinical groups were differentiated from
normals by smaller social support systems and the

receipt of support from others rather than the

an grace

22193» a
1‘1 socia.

”p.310?
IU‘MOIB b

"i". ii I

l

"N U 5:

l.-

113161.53]
(LS 'n'as r

In.

""5; he

17

reciprocal interchange of support with others. Brown,
Birley, and Wing (1972), in discussing the composition
of social networks, evaluated the relative merit of kin
contacts among schizophrenic individuals. They report a
curvilinear relationship between support and psychiatric
status. No contact or isolation from kin and exclusive
reliance on kin in a highly emotional atmosphere were
both detrimental to the psychiatric status of former
patients, whereas a more moderate amount of contact with
kin was related to better status.. In regard to the
social networks of mothers with severe mental disorders,
Cohler, Grunebaum, Weiss, Hartman, and Gallant (1976)
have shown that less adaptive maternal attitudes toward
child care were related to poorer functioning (less
contact, less closeness) in a number of social roles
including relationships with friends, spouse, and
parents. Among less disturbed parents, social support
moderates depression through its influence on feelings
of self efficacy in mothers of difficult infants
(Cutrona & Troutman, 1986).

A number of recent studies illustrate the
importance of social support in the parenting context.
For example, Crockenberg (1981) has shown that for
mothers of infants, social support influences maternal
responsiveness and is a strong predictor of the security

of the mother-infant attachment that develops. Among

3227.855
s;::crt1

a AI ‘
Z2 ,‘JJCJ

2:17.853

1
"A
. t

....al 3

n".. ‘
a ‘
1“qu

u 1“
l. '. h
.n‘
a
:v.‘
i‘-

IIIV‘

“3% at
“e

5.‘ .

. \Cns

‘IIL. ‘

18

mothers of older children, limited contact with
supportive others and social isolation have been related
to problematic parenting {e.g., aversive parenting in
mothers of Oppositional children (Wahler, 1980: Wahler,
Leske, & Rogers, 1979) and child abuse (Salzinger,
Kaplan, & Artemyeff, 1983)}. Cochran and Brassard
(1979) discuss a number of mechanisms through which
social support networks may influence parenting. Social
networks provide access to both emotional support and
practical assistance: these may enhance the mother's
responsiveness to her children and reduce stress in the
caregiving role. Social networks may also be a source
of stress if the network places demands on the mother
which divert her resources from parenting (Belle,l982)
or if network members pressure the mother to rear her
children in ways they feel are appropriate. The social
network may influence parenting by providing
child-rearing models and controls, providing new
solutions for dealing with children, encouraging
particular patterns of parent-child interaction, and
placing sanctions on the type of parental behavior
exhibited. Isolated mothers would not benefit from
these network influences and may be likely to use the
style of parenting they experienced as children. This
may contribute to the intergenerational transmission of

deviant parenting styles.

19

The structure or composition of the support
network, as well as the support it provides, also may
influence the course and effectiveness of parenting.
Hammer, Gutwirth, and Phillips (1982) indicate that
being a parent is likely to alter the type of network
one has. They found that mothers, as compared to
nonmothers, had about the same size social network, but
mothers had fewer average daily contacts with adults and
a larger proportion of their networks was composed of
kin. Thus, for mothers with severe disorders, their
parent status should promote greater contact with kin,
but an overreliance on kin may be a risk to their mental
health (Brown, Birley, & Wing, 1972). A balancing of
contact with relatives and other individuals (friends,
neighbors) may be more beneficial to the mother. Dell
and Appelbaum (1977) describe a particular pattern of
trigenerational enmeshment in which single parent
mothers return to their family of origin to live with
their children. This living arrangement may have
detrimental effects on both the mother and the children
especially if the mother had not been able to adequately
differentiate from her family of origin in the first
place. This may be particularly true of mothers with
severe disorders and is more applicable if the mother is

older.

The presence of a spouse or significant other can

ze an it
taregivi

:ype of

 

I."‘

a!

uvo .
.. ,. 1
55.531

‘ a
stzszac

attache

USZEES

”IR-n
:::.|iclc

'1

‘.:q‘

.‘1 l'n

‘. Be

H,

u ‘
9. mi

3?

‘-x

20

be an important source of support for the mother in her
caregiving role and has substantial influence on the
type of parenting behavior the mother exhibits (Lamb,
1981). Levitt, Weber, and Clark (1986) have shown that
spousal support influences maternal affect and life
satisfaction which, in turn, related to security of
attachment between mothers and their infants. Further,
a negative relationship with her husband or boyfriend
fosters stress in the mother's life, detracting from her
psychological adjustment and available personal
resources for parenting. Poor marital relations have a
negative influence on children's development (Maccoby &
Martin, 1983) and can lead to a variety of dysfunctional
family patterns, such as triangulation of the child in
the marital difficulties (Bowen, 1978).

The quality of support the network is able to
provide will vary according to the social context in
which it occurs. Belle (1982, 1984) has emphasized that
the supporters in the social networks of low-income
individuals are likely to stuggle with their own limited
resources, attentuating the support they are able to
give. Thus, the resources of the physical living
environment may mediate the influences on parenting that

social support networks provide.

:,....
sale

ulifli
[in]

IAO‘p
sepii

I! c l
"51

(ll
--

.
a. ‘h
" in.

IA '
‘ A‘

3

OJ...

 

'5 :74.
5.1

 

21

Physical Living Environment. The physical
environment, while related to the social network in
which the mother exists, may also be related to the
mother's well-being and her parenting. Adequacy of
housing, comfort with the neighborhood, access to
transportation and shopping, etc. influence the degree
of stress a mother experiences in day-to-day living. If
stressful, these influences undermine the personal
resources the mother has available for parenting. Many
of these influences are likely to be a function of
social class (Belle, 1982: Brown & Harris, 1978), and
many mentally disordered mothers are of low
socioeconomic status, and thereby in a more stressful
environment. Indicative of the strains on parenting and
the impact on children of social disadvantage are the
findings from the high risk research of Sameroff,
Seifer, and Zax (1982). In their prospective
longitudinal study, they found that low social class was
more predictive of problems in the children than was the
mother's diagnosis (of schizOphrenia or affective
disorder) alone. Having a difficult child, regardless
of social class, places added stress on one's parenting

abilities.

22

Child Characteristics

Finally, individual child characteristics are
likely to influence the type of parenting mothers
evidence. Bell's classic work (Bell, 1968: Bell and
Harper, 1977) has shown that different types of children
demand different responses from parents. In fact,
children may condition different parenting responses in
their parents as a result of their characteristics.
Thomas, Chess, and Birch's (1968) work on variations in
temperament underscores this concept. Patterson (1980:
1986) has related maternal rejection to the greater
frequency with which a mother of a deviant child
experiences ”daily hassles" involving her problem child
and to the mother's sense of failure in her parenting
responsibilities as a result of the child's maladaptive
behavior. Of course, parents also directly influence
the behavior of their children, and the process of
influence is likely to be mutual and transactional over
time as characterized by Sameroff and Chandler (1975).
Research Goals

As the preceding discussion indicates, the
potential determinants of parenting, especially in women
with severe mental disorders, are diverse and complex.
The present investigation examined a number of these
determinants in order to explore individual differences

in parenting attitudes and practices among mothers with

23

severe mental disorders. As Belsky (1984) has
suggested, it is essential to move beyond univariate
approaches for understanding parenting and move toward
more complex integrations of variables at a number of
levels. Following Belsky's (1984) framework, the
present research assessed multiple aspects of the
following principal domains: the mother's developmental
history (e.g., history of separation and loss, perceived
parenting experienced), personality (e.g., severity and
chronicity of disorder, self esteem), social context
(e.g., social support network, demographic
characteristics, physical living environment), and
child's individual characteristics (extent of child
behavior problems). Each of these domains was eXplored
for descriptive purposes to characterize in detail the
current context in which severely disordered mothers
perform their maternal role. Further, a number of
features were examined for differences among the mothers
which would be likely to enhance or detract from their
parenting abilities. More specifically, multiple
regressions were conducted to explore predictors of age
at first birth, self esteem, psychological distress,
extent of perceived child behavior problems, and a
number of measures of parenting attitudes. Potential
predictors of these factors were drawn from the various

domains assessed.

24

In order to organize and integrate the multiple
determinants of parenting for severely disordered
mothers, a conceptual model, adapted from Belsky's
(1984) scheme, was initially formulated. This model is
presented in Figure 2. Components of the model include
the mother's developmental history, severity and
chronicity of her disorder, self esteem, social support,
socioeconomic status, quality of the physical living
environment, and child behavior problems as these
factors related to parenting style. In comparison with
Belsky's conceptual model, marital relations and work
are notably absent. While these areas are influential
determinants of parenting in general, severely
disordered mothers are often not married and are usually
unemployed, and thus these components were excluded.
Further, based on the evidence of intergenerational
continuity in parenting discussed previously, a direct
link from developmental history to parenting was
proposed, rather than restricting the influence of
developmental history to its single effect on parenting
as in Belsky's model.

Evaluation of the model proceeded in two main steps
based on Scarr's (1985) discussion of proximal and
distal influences on behavior. Scarr noted that a
predominant bias exists in much of current psychological

theory and intervention efforts which gives preference

25

 

 

wedaeenem

 

msodponm

Hoa>snom
dawno

 

         

  

pros
icowa>cm
med>eg
Assam» .

  

  

swerve:

paommsm
Hmnoom

 

 

amoeba
mamm

assodeoneo
s
hvfiho>om

 

sonoanoasa aaeae>om animawsaenem mo mensesaneaon mo Heeoz seam

.N ouswah

26

to examining proximal factors (i.e., influences in the
immediate context) as the primary determinants of
behavior. However, consideration of distal factors may
diminish the explanatory power of proximal variables.
In the current study, the first step involved
examination of only proximal influences on parenting
behavior. In the second step, distal variables (i.e.,
features of the mother's developmental history) were
added to the model to examine the extent to which they
altered the initial relationships in predicting
parenting style.

The use of path analysis requires that the model
tested is recursive (i.e., a unidirectional causal flow
between variables.) However, in the proposed model
bidirectional influences are posited between a number of
the proximal variables. For example, a bidirectional
influence is noted between self esteem and social
support. One may hypothesize that higher self esteem
predicts higher levels of social support because a
mother who is more accepting of herself is more likely
to attract more supportive individuals into her social
network. Conversely, high levels of social support may
predict higher self esteem because more assistance and
caring from others is likely to augment one's own
feelings of self worth. Both explanations are

plausible. Therefore, separate models will be examined

27

to evaluate if causal flow in one direction fits the
obtained data better than causal flow posited in the
opposite direction (e.g., self esteem predicting social
support vs. social support predicting self esteem).

An advantage of path analysis is that it allows for
correlations between variables to be decomposed into
component parts, namely direct, indirect, and spurious
influences. As a result, a more detailed understanding
of one variable's influence on another is attained. In
the present model, the strength of direct effects
between the various predictor variables and parenting
style were compared to ascertain which factors most
powerfully influence parenting. The indirect effects of
predictor variables on parenting also were examined and
compared to their respective direct effects. Belsky's
(1984) conceptualization would predict that the
developmental and personality variables would have the
strongest impact on parenting, followed by the social
contextual variables and then the individual child
characteristics. The accuracy of this relative
weighting was evaluated.

The basic hypothesis tested was the adequacy of the
conceptual model to depict relationships in the data in
the prediction of parenting style. More specific
hypotheses are contained within the model and are

delineated below:

28

The mother's developmental history (history of
separation and loss, perceived mothering received as
a child) directly affects parenting style negatively
and indirectly affects parenting style through its
influence on personality (self esteem and severity
and chronicity).

The mother's personality (self esteem, chronicity and
severity of disorder) directly affects parenting
style (positively for self esteem, negatively for
chronicity, severity of disorder) and indirectly
affects parenting style through its influence on
social support.

The adequacy of the mother's social support system
directly affects parenting style positively.
(Indirect influences on parenting may also result
from social support's direct influence on
personality.)

Socioeconomic status directly affects positively the
quality of the physical living environment and thus
indirectly affects parenting. The adequacy of the
physical living environment directly affects
parenting style positively.

The child's perceived psychological difficulties
directly affect parenting style negatively.

Although the present model posited specific

relationships between the variables, attempts were made

29

to improve the fit of the model to the data through the
addition and deletion of alternate paths. Alterations
of the model were based on conceptual grounds as well as

the strength of observed relationships in the data.

51.. NH
:15 Ebb-
"

”h

511‘

i

:32 of .

VA >'n H"

-'UrbI|V ‘.
n'. ..

".n5:n

..,
mo.

“beerEI

l
If. .

I
"‘Wi‘c t
‘ .

METHOD

Subjects

The participants were 48 mothers hospitalized in
one of four state of Michigan psychiatric facilities,
Northville (n=35), Kalamazoo (n=4), Ypsilanti (n=3), and
Clinton Valley (n=6). (Interviews were conducted at two
different time periods at Northville, the largest state
psychiatric center, in order to reach the obtained
sample size.) In each of the facilities, social work
staff members on the inpatient units identified patients
who were mothers with at least one child under age 13
with whom they had been living prior to hospitalization.
Among these patients, the social workers also screened
from participation those patients whose current mental
status was too unstable (e.g., those in seclusion) for
them to be interviewed.

All potential participants then were approached and
asked for their permission to be interviewed. Sixty-
seven percent of the potential participants agreed to
take part in the study. The mothers who refused to
Participate were compared to the participants on a
number of features and were not found to differ
Significantly in regard to age, race, marital status,

education, diagnosis, total number of hospitalizations,

30

pacer 0f
age at H

The
generally
socisecon
greicmina:
agecifica.‘
isllingshe
status. F
The partic
:izth grac
Sign sch .
graiuate.
mm
31.4 \{21

51114“

31

number of hospitalizations in the past three years, or
age at first hospitalization.

The demographic characteristics of the participants
generally indicated that the sample was of low
socioeconomic status, unemployed, poorly educated,
predominantly of minority race, and not married. More
specifically, 71% of the sample were at or below 20 on
Hollingshead's four-factor scale of socioeconomic
status. Previous to admission, 88% had been unemployed.
The participants varied in educational level as follows:
ninth grade or less, 13%: tenth or eleventh grade, 27%:
high school graduate: 31%: part college, 23%: college
graduate, 4%: and graduate degree, 2%). The age of the
participants ranged from 17 to 50 with a mean age of
31.4 years. Sixty-five percent of the participants were
black, and the remainder were white. Although 21% of
the participants currently were married, the large
majority were not married (never married, 52%: separated
or divorced, 27%). The mean number of children of the
participants was 2.1 (range: 1 to 6): 46% had one child.
In comparison to a survey of inpatient mothers in all
state of Michigan psychiatric facilities (Rogosch &
Mowbray, 1987), the present sample differed in that it
consisted of 9% more individuals of minority race and
14% more never married women.

The major diagnostic groupings of the patients

I l
n-rb‘ 1
J.“ J
I

gsychie
6.8 is:
aizissi
seven in

:t' the

h . 0r
£3,951};
;‘

32

based on hospital diagnosis were as follows: 65%

schizOphrenic disorders, 13% schizoaffective disorder,
and 21% affective disorders. Ninety-one percent of the

participants previously had been admitted to a

psychiatric facility: the mean number of admissions was

6.8 (sd=5.8, range: 1 to 24). The number of psychiatric

admissions in the last three years ranged from one to

seven with a mean of 3.0 (sd=l.7, range: 1 to 7). Most

<>f the participants had been admitted on court order for
60- or 90-day hospitalizations. At the time of the

.interviews, the median length of stay for the sample was

61m5 days.

Laterials

The measures used in the research interview
<:onsisted of both standardized instruments and
<3uestionnaires devised specifically for this study.

liach of the measures is described below in the order of

£>resentation during the interview.
Current Life Situation Questionnaire. This

xneasure, designed for this study, assessed a variety of

<3emographic characteristics of the participant as well

its physical features of and satisfaction with her

£>rehospital home environment. Demographic information

assessed included age, birthdate, race, marital status

Elnd history, education, prehospitalization employment,

and educ
3;;1ica‘:
estacli:
.1375) c
were det
nestior

the care

Pfemanc
mental
Latencie
asked at

“Eiiinc

33

and eductation and employment of her spouse, if
applicable. (Education and employment were used to
establish socioeconomic status using Hollingshead's
(1975) criteria.) Income source and satisfaction also
were determined. The participant was asked a number of
questions about her children: their sex and birthdates,
who cared for them while the participant was in the
hospital, who was the father of each child, whether each
pregnancy was planned, any physical problems during her
pregnancies, and psychiatric problems or
hospitalizations associated with the pregnancies or
infancies of her children. Finally, the mother was
iasked about her household composition (all persons
lfesiding there and relationship to mother), housing type
(apartment, house, etc.: rented, owned, relative's),
lliving situation adequacy, length of residence in
EDrehospitalization living situation, and satisfaction
‘Vith housing and neighborhood, and in-town
tzransportation availability. (See Appendix A for a copy

<>f this instrument).

Early Life Questionnaire. This measure, devised

 

Ifor this study, included questions concerning with whom
tzhe participant lived during different periods of her
<=hildhood. The occurrence of separations from her
another and father was assessed as well as the reasons

for such separations. The subject was asked to rate her

34

relationship with different family members as a child
and to rate her perception of the marital adjustment of
her parents. The participant was asked if she felt
close to anyone as a child and who those persons were.
Questions also specifically addressed the presence of
psychiatric disorder in other family members. (See
Appendix B for a copy of this instrument.)

Children's Reports of Parental Behaviors Inventory
(QBEBI). (Raskin, Booth, Reatig, Schulterbrandt, &
(Ddle, 1971.) This questionnaire is an abbreviated
‘18-item version of a scale originally developed by
£3chaefer (1963) in which a subject's perceptions of
[parent's behavior toward him/her as a child are
Eissessed. Factor analyses conducted by Raskin et a1.
(1971) have yielded three major dimensions, positive
janolvement, negative control, and lax discipline. The
Eibbreviated version consists of the 48 items that loaded
laighly on one of the three major dimensions. The
£>articipants in this study were asked to report only on
tzhe behavior of their mothers. Sample items for each of
tzhe three dimensions were as follows: positive
i.nvolvement, ”Comforted me when I was afraid," "Always
J.istened to my ideas and opinions:" negative control,
"Said that someday I'd be punished for my bad behavior,”
"Told me how much she had suffered for me:" lax

(Siscipline, ”Seldom insisted I do anything," "Didn't pay

35

much attention to my misbehavior." Response options for
the items were ”True of my mother,” "Sort of true of my
mother,“ and “Not true of my mother." Each item was
scored from one to three depending on the level of
agreement: mean scores were then obtained for each
dimension. Internal reliability coefficients
(Cronbach's alpha) for the three subscales were .90,

.88, .67 for positive involvement, negative control, and
lax discipline, respectively.

Marked differences on this scale between depressed
.inpatients and normal controls have been reported by
(Zrook, Raskin, and Eliot (1981). Despite the
lcetrospective nature of the scale and the issue of
Ireconstruction in adulthood, Schaefer and Bayley (1967)
laave reported significant correlations between adult
tsubjects' retrospective accounts (obtained with the
<2RPBI) and parents' reports of their own behavior
<3btained when the subjects were aged 9 to 14. (See
Ikppendix C for a copy of the 48-item version of this
instrument.)

Sympton Checklist - 10. (Nguyen, Attkisson, &
EStegner, 1983). The purpose of this instrument is to
<>btain brief assessment of the participant's perception
<>f the severity of psychological distress. The SCL-lO
5L3 an abbreviated version of the original SOL-90

(:Derogatis, Lipman, and Covi, 1973: Derogatis, 1975) and

isbased
the majo
535-90 (1
include.
interest
3 feelil
five-poi:
all.‘ T r
shéy we:
Estrume:

h

(I)
(D

I

m¥elgpe<
Satisfiec

36

is based on the ten items with the highest loadings on
the major factors derived by factor analysis of the
SOL-9O (Hoffman and Overall, 1978). Sample items
include, ”How much were you distressed by feeling no
interest in things?" and ”How much were you distressed
by feeling tense or keyed up?” Subjects respond on a
five-point scale ranging from "extemely" to "not at
all.” The internal reliability of the scale for this
study was .82. (See Appendix D for a copy of this
instrument.)

Self Esteem Scale. This lO-item questionnaire
cieveloped by Rosenberg (1965) is a global measure of
eself-esteem. Sample items include, "On the whole I am
asatisfied with myself," and "At times I think I am no
Eyood at all.” Response options are on a four-point
£3cale from strongly agree to strongly disagree. Based
<>n Rosenberg's 6-scale Guttman scoring of the
<3uestionnaire, internal reliability for this study was
..62. (See Appendix E for a copy of this instrument.)

Social Support Questionnaire. The participant's
Esocial support network was investigated using a modified
‘rersion of the Social Support Questionnaire (SSQ)
(Seveloped by Bogat et al. (1983). The modified
QIuestionnaire has a similar format to the original 880:
lb"lowever, fewer questions are asked (16 vs. 9), and for

some questions, the wording of the items differs,

although
:cc'ified
::nfi:’e i
feel upse
:ziliren,
is sick,
:nild?'),
6:: pravi
“4.“.

......a:ed

ascial su

fizzlcnal

ital both

-—,

"Q‘a

37

although the basic content is similar. For each of the
modified version's nine questions (e.g., "Who can you
confide in when you have a personal problem or when you
feel upset?" and "Who helps you with taking care of the
children, such as babysitting, helping out if your child
is sick, helping if an emergency occurred with your
child?”), the respondent lists the names of individuals
who provide her with the type of social support
indicated in each question. Five types of positive
social support are assessed with two questions each:
<:ompanionship, instrumental aid, information and advice,
eamotional support, and child care assistance. One item
Ifrom both the instrumental aid category and the
information and advice category dealt with content
Especific to child care, and these items comprised the
<2hild assistance category. The ninth question inquired
enbout individuals who were stressful members of the
nnother's social network. All of the individuals named
13y the participant were listed on a separate page, and
tzhe participant was asked to provide further information
Elbout each one: gender, relationship to the participant,
£>arental status, frequency of contact with the
£>articipant, and satisfaction with the relationship.
The following variables were computed from the
u'lodified 580: number of total supporters, number of

Supporters providing each of the types of support

‘JIH
("

nr‘
.164

"in

qr}
vs\

a._

‘4‘

'iU,.

I
. va
.9:

38

(companionship, information, aid, emotional support,
childcare assistance, and stress), number of kin and
nonkin supporters, the presence of specific individuals
providing support (i.e., parents, spouse, children,
etc.), number of supporters seen weekly, and number of
supporters with whom the participant was satisified.
(See Appendix F for a copy of this instrument.)
Mother-Child Relationship Evaluation. (Roth,
1980). The MCRB contains 48 items assessing maternal
attitudes and perceptions of what constitutes
.appropriate qualities of the relationship between mother
aand child. The subject responds on a five-point scale
:from ”Strongly Agree” to "Strongly Disagree." The
.ihstrument is scored for four subscales: Acceptance,
(Dverprotection, Overindulgence, and Rejection.
lieliability and validity indices are reported by the
iauthor: however, these indices were based on a small,
laomogeneous sample of mothers. (See Appendix G for a

<:opy of this instrument.)

Subsequent to data collection, the subscales of the
DiCRE were evaluated for their psychometric properties.
Iklphas obtained for the four subscales were as follows:
,.69, .78, .67, and .47, for Acceptance, Overprotection,
(Dverindulgence, and Rejection, respectively. In order
1:0 assess the extent to which the four subscales were

Uneasuring distinct aspects of the mother-child

relatio

:safim

:‘iffere
:eiongel
fame 1
:cr: la'
finding:
the ECR!
cznstruc

In
”53 itc
Princi;
afiaax
Haluat,
Vim {81
lire Bu!

“31'131

1m th.

39

relationship, the items of the MCRE were subjected to a
confirmatory factor analysis of the four subscales.
Half of the items (2324) correlated more strongly with a
different subscale than the subscale to which they
belonged. Only the Overprotection subscale retained a
degree of distinctness: however, four of its items also
correlated more strongly with other subscales. These
findings indicated that the subscales, as assessed by
the MCRB scoring, were not tapping unified and distinct
constructs.

In order to obtain more distinct dimensions, the
IHCRE items were subjected to exploratory factor analysis
(principal components factor analysis followed by
‘varimax rotation). A number of solutions were
evaluated. From the five factor solution, the items
‘with factor loadings above .40 on the first two factors
‘dere subsequently subjected to a confirmatory factor
analysis. The results of this analysis demonstrated
‘that the items of each of the two new subscales
<:orrelated more strongly with their own subscale than
‘with the other subscale. Alphas computed on these two
Iaew subscales were .87 and .83. The two subscales were
<:orrelated (r =.26, p < .05). The content of these two
Inew subscales was examined. The first subscale retained
:Eive of the items of the original Overprotection

asubscale with additional items tapping overcontrol and

Co a u .. . a-“ W1.
8 N b «“5 Pk A V
'O ' V! i Q, U I I

E

a...
v8.
d
.‘jl
OvV
fl";-
a...“

no

restriction of autonomy. Sample items are as follows:
"Children should always be kept calm," “A mother should
defend her child from criticism,” and "A mother should
control her child's emotions.” This subscale,
containing 13 items, was labeled ”Overprotection-
Overcontrol' (OVPOVC). The second subscale contained
items that conveyed a sense of hostility on the mother's
part toward the child and treatment of the child as an
object with little awareness of the child as a feeling
person (e.g., “Children are like small animals and can
be trained the same as puppies," "Young children, like
toys, are for their parents' amusement,” and "My child
annoys me.”). This subscale of 10 items was labeled
"Hostile Objectification" (HOSOBJ).

Parental Attitude Survey Scales (PAS). (Hereford,
1963). To assess the participant's sense of competence
and/or frustration in the maternal role, 14 items with
the highest factor loadings on the difficulty/confidence
subscale (PASD) of the PAS were administered to the
participants. Sample items included, "I feel I am faced
with more problems than most parent," and "A parent has
to suffer much and say little.” Subjects responded on a
four-point scale from "strongly agree" to "strongly
disagree.” (See Appendix H for a copy of these items).
Subsequent to data collection, reliability of the scale

was assessed (alpha =.87). Two of the items, however,

xrrela
The re:

m ccm;

fituati
dialogu
unage
aprsac
hciall
milec:
Ewes:
Fiftici
.598 AP
Fa
S{S'llak

ather'

41

correlated negligibly with the scale and were deleted.
The remaining 12 items had an alpha of .84 and were used
to compute scores for the scale.

Sensitivity to Children Questionnaire. (Stollak,
Scholom, Kallman, & Saturansky, 1973). The STC presents
the mother with a series of problematic parent-child
situations, and the subject is required to describe in
dialogue form how she would respond to the child to
manage the situation. This active, less structured
approach is thought to be less subject to selection of
socially appropriate responses and may more adequately
reflect the mother's parenting style. A subsample of
five STC situations were administered to the
participant, and her responses were recorded verbatim.
(See Appendix I for a copy of this instrument.)

Factor analyses of responses (Teyber, Messe, &
Stollak, 1977) indicate several dimensions on which the
mother's responses can be scored, one of which is an
acceptance (moving toward) vs. rejecting (moving
against) dimension. This approach was chosen and
elaborated on in order to gain a global perspective on
the participant's parenting style. Baumrind (1967,
1971) has conceptualized different styles of parenting
and emphasized the importance of providing structure and
control in addition to acceptance vs. rejection.

Maccoby and Martin (1983) expand on the work of Baumrind

and 0th!
scheme 4
figure E
msgons:
garent-c
mienanc
ii: nsic
nfulger
Baumrinc
fii‘zided

Thj
firtici;
each of
Fiftici;
lifePQQC‘
78 fun

“533m

.,
“latte

42

and others and present a two-dimensional classification
scheme of parenting styles. This scheme is presented in
Figure 3. The two major dimensions are accepting,
responsive, child-centered vs. rejecting, unresponsive,
parent-centered and demanding, controlling vs.
undemanding, low in control attempts. These two
dimensions yield four parenting styles: authoritative,
indulgent, authoritarian, and neglecting. Thus,
Baumrind's original permissive parenting style has been
divided into indulgent vs. neglecting.

This conceptual organization was used to score the
participants' STC responses. Initially, responses to
each of vignettes were aggregated across the
participants. For each of the vignettes, two
independent raters categorized each solution into one of
the four parenting styles which most closely
characterized the style of the response. An additional
fifth category was also required for those responses
which were characterized by psychotic process and were
unanalyzable based on the four categories alone.
Interrater agreement for the coding of the responses as
assessed by Cohen's (1968) weighted kappa was .87. (See
Appendix I for examples of responses coded in each of
the parenting categories for each vignette and for
frequencies of each parenting category coded for each

vignette.) In addition to coding for parenting style,

‘Fu'x'n'

inunu‘Q ‘

35"!“ A?
, d”
‘W'aubl

”V." \ \
.J.;fin_

" L.' Y"
'3’” in
)F‘flr'
a. Own

3"‘a

\-

SISTY]

43

Figure 3.

Conceptual Organization of Parenting Styles

 

 

 

ACCEPTING REJECTING
RESPONSIVE UNRBSPONSIVE
CHILD-CENTERED PARENT-CENTERED
\
Authoritative- Authoritarian
DEMANDING - Reciprocal (Power -
CONTROLLING (High in Bi- Assertive)

Directional<x
Communication)\‘\

 

 

 

 

 

 

 

 

\
\
\.
\ § 0
Indulgent Neglecting
UNDEMANDING - (Ignoring,
LOW IN CONTROL Indifferent,
ATTEMPTS Uninvolved) \
Psychotic-
Intrusion

 

Ngt . Diagonal corresponds to dimension assessed by
STCSTYLE.

:lze code
involve<

ms stre

'13
0

teams:

”5?~ns<
“sleet

fish in:

«335'

44

the coders also scored whether or not the solution
involved the use of physical punishment. As this code
was straightforward, the interrater agreement was 100%.
Following the coding of the solutions, the
responses were reaggregated by participant, and the
number of responses in each of the parenting style
categories was determined for each participant. In
order to condense the information into one dimension to
use in data analysis, a parenting style dimension was
derived. As authoritative parenting is regarded as the
most beneficial parenting style in that it incorporates
both of the goals of acceptance and providing structure,
a score of "l” was given for each authoritative
response. Both indulgent and authoritarian parenting
neglect one of the important goals of parenting, and
each indulgent or authoritative response was given a
score of ”0'. Finally, neglecting parenting attends to
neither goal of acceptance or providing structure, and
each neglecting response was given a score of "-1”.
Similary, psychotic responses were also scored as "-1"
in that attending to either goal of parenting is not
possible if responding in a psychologically disordered
manner. This parenting style dimension thus represents
a diagonal across the schematic depiction presented in
Figure 3. Each participant's score was the sum of

scores across the five vignettes. A reliability

assessme
algha of
Eielbrcc
:xild be
:zternal
withdraw
:zsturi
iteract

Iruelty)

In”
ulL-yeten
426, T

a“ v
.‘ .

R
'a.n‘~.

.
I ‘

MIA}
‘SKI '

45

assessment based on this coding scheme resulted in an
alpha of .61 for the five STC vignettes.

Child Behavior Checklist (CBCL). (Achenbach &
Edelbrock, 1983: Achenbach, 1986). The CBCL assesses
child behavior problems in two broad behavioral domains,
internalizing behavior (e.g., depression, social
withdrawal, schizoid tendencies, somatic complaints,
immaturity), and externalizing behavior (e.g.,
hyperactivity, sex problems, delinquency, aggression,
cruelty). Additionally, an assessment of social
competence in activities, school, and social behavior is
made. The CBCL has been standardized on over 3600
clinic- and nonclinic-referred children. Normative data
have been developed separtately for boys and for girls
in the following age groups: 4—5, 6-11, 12-16. An
additional version of the CBCL has been constructed for
children aged 2 to 3 (Achenbach, 1986). If the
participant had more than one child under age 13, she
was asked to select one of her children with whom she
had most concern and complete the CBCL on this child.
(See Appendix J for a copy of these instruments.)

Qbserver Checklist. (Fishman, 1981). Following
the completion of the interview with the patient, the
interviewer completed a checklist of his/her perceptions
of the participant's behavior during the interview. The

checklist includes a list of 69 behavioral items

:niicati‘
42affec1
feelings
::;y of 1
:2: rviee
1.1 srder
:zgnt be
assessed
itequenq
:‘xther,
Zighly s:
Fifticipe
far fart?
implies
literviee
'5 the it

'7‘ 5' .
fission:

46

indicative of psychopathogolgy. Additionally, a list of
42 affects provides an indication of the participant's
feelings during the interview. (See Appendix K for a
copy of this instrument.) Thirteen (27%) of the
interviews were conducted with two interviewers present
in order that interrater agreement for the checklist
might be assessed. Unfortunately, many of the items
assessed behaviors that occurred with very low
frequency, and interviewers rarely endorse these.
Further, many of the items were not applicable to the
highly structured interviews conducted with the
participants. As a result, only one index was retained
for further consideration: this item measured the
interviewer's confidence in the validity of the
interview. The interrater agreement was 92%. Over all
of the interviews, eight were considered to be of
questionable validity: the participants in these
interviews appeared to be more disturbed rather than
being resistant to the interview. The interviewers
noted which parts of the interview were questionable,
but these parts varied across the participants.

gage Record Review. In addition to the information
obtained in the patient interview, the patient's
hospital record also was reviewed by the interviewer to
obtain demographic characteristics, hospitalization

history, and information regarding the current hospital

2.126”

s u
.Vfifinn
e.uu.C a
—

.‘ : .;
niluC

u. e5.
5,‘c
3 .
‘H ‘n
§l“
:n
'1

4?

admission and treatment. Where possible, the
interviewer also confirmed information provided by the
participant in regard to current life circumstances and

historical details of the participant's life.

Procedure

Interviewers in this study consisted of advanced
clinical psychology graduate students who had received
training in administering the specific research
instruments used in the investigation. The interviewer
met individually with the participant in a private area
on the hospital ward. Initially, the interviewer
determined that the participant was oriented to person,
place, and time before proceeding. The interviewer
discussed the content of the interview, assured the
participant of confidentiality and anonymity, and
reviewed an informed consent form with the participant.
If the patient chose to participate, she then signed the
consent form. (See Appendix L for a copy of the consent
form.) Additionally, the participant was asked for her
consent to contact the person caring for her chi1d(ren)
so that the caregiver could answer questions pertaining
to the child's adjustment. If the participant agreed,
she signed an additional consent form and provided the
address of the caregiver. (See Appendix L for a copy of

this consent form.) The participant was not required to

:1“ t

:38 ta

3‘ the

Laterv

EEfSOr
5956'):

15‘

\"W3!

9.
‘4r ,
‘~
5
“ E

48

give this consent in order to participate in the study,
and 52% declined to give consent.

The interview consisted of the completion of the
structured interview questionnaires and the rating
scales. The items of the rating scale instruments were
read aloud to each participant: the participant was
given a corresponding copy of the response options for
the rating scale items, and her responses were recorded
by the interviewer. The length of time to complete the
interview ranged from one to four hours (fl = 2.0). Care
was taken to adjust the demands of the interview to the
participant's ability to maintain attention and comfort
with the questions. Breaks were taken when appropriate,
and if needed, the interview was completed on the
following day.

If the participant had given consent to contact the
person caring for her children, a letter was sent to the
caregiver requesting that he/she complete the Child
Behavior Checklist for the participant's child or
contact the experimenter if further information was
desired. The CBCL and an informed consent form for the
caregiver (see Appendix L) were sent with the letter.
The experimenter was restricted from contacting the
caregiver directly unless the consent form was returned.
The return rate of the caregiver consent forms and

CBCL's was very poor: only five were returned. Given

49

this low rate of return and the large number of
participants who had refused permission to contact the
caregiver, the information obtained from completed child

questionnaires by caregivers was not considered further.

.I

3C

I

.0

IIU
l3.

RESULTS

Experience in Family of Origin

Instability in the participants' families of origin
was frequent. Less than half of the participants (48%)
had lived with both of their parents while growing up.
Black participants were less likely than white
participants to have lived with both parents {£11,47) =
4.67, g < .05: 35% vs. 71%}. Forty-eight percent of the
participants had experienced at least one divorce or
separation of their parents, and 31% had experienced at
least one remarriage of the custodial parent.

Separation and loss also were significant issues in
a number of the participants' lives. Thirty percent had
experienced at least one separation from their mothers
before age 13. Reasons for separation included
placement of the participant in foster care, being sent
to live with relatives, hospitalization of the
participant's mother, etc. Six of the participants
(12.5%) experienced loss of their mothers either by
death or abandonment. Separation from the participants'
fathers was even more prevalent: 58% of the participants
had experienced at least one separation from their
fathers before age 13.

Most of the participants (88%) reported that there

50

u“"
MV

..5
Id“

\HH

51

was someone to whom they felt close while growing up.
However, only 31% of the participants' mothers and 17%
of the participants' fathers were included as
individuals to whom the participant felt close.
Siblings and other relatives were frequent choices of
individuals with whom a close bond existed.

The participants also characterized their
experiences of their own mothers through completion of
the abbreviated 48-item version of the CRPBI. The means
and standard deviations for the three scales of positive
involvement (PIM), negative control (NCM), and lax
discipline (LDM) were as follows: 2.46 (gg = .45), 2.14
(gg = .54), and 1.87 (gg = .48). Raskin et al. (1971)
reported scale means for a normal sample and two samples
of patients hospitalized with depression. However,
standard deviations were not reported for the respective
scale means, restricting the comparison with the present
results. Based on the means, the current participants
reported as much positive involvement with their mothers
as did Raskin et al.'s normal sample and more positive
involvement than indicated by the depressed samples.
Further, the current participants tended to report more
negative control and more lax discipline on the average

than did both the normal and depressed clinical samples.

p
§

H
\-
9“"
hV-a
h
l
a.
I

1

II.
In.

E

A

in.
U

'0‘

:3

V“

I

52

Influences on the Onset of Parenthood

Several factors suggest that many of the
participants had not been well prepared for the onset of
parenthood and the concomitant responsibilities of child
rearing. The average age among the mothers at first
pregnancy was 20.4 years (range: 11 - 36). The mean age
at which the participants gave birth to their first
child was 21.1. Thirty-six percent of the mothers had
become pregnant and 29% had given birth to their first
child before age 18. A multiple regression was
conducted with age at first birth as the dependent
variable and diagnosis (affective vs. schizophrenic
disorder), education, and age at first hospitalization
as predictor variables. (See Table l.) Schizophrenic
diagnosis, less education, and earlier age of first
psychiatric hospitalization related to giving birth at a
younger age, and together accounted for 49% of the
variance.

Many of the participants (69%) had given birth to a
child without being married. Fifty-two percent of the
participants have never married: therefore, the
incidence of single parenthood is likely to be
persistent. Sixty-one percent of the participants'
children were reported to be unplanned, and 52% of the

participants reported that none of their children had

53

Table 1.

Multiple Regression to Predict Age at First Birth

2
Step Variable ‘5 to p_ MR .A§_ Final Step
Enter /3 p
l. DX 17.29 .000 .54 .29 -.43 .001
2. EDUC 13.52 .002 .68 .17 .42 .001
3. AGEHOSPl 6.35 .026 .73 .07 .28 .026
32 = .53
Ad]. 32 = .49

5 (3,41) = 15.59, P. < .001

J

vb

n r6
,3.L

,.
Hut

0..

.1.
.11..
lo.‘

5:31

A
l‘rv

v...

54

resulted from planned pregnancies. Psychiatric
hospitalization occurred during a pregnancy for 34% of
the participants. Forty-three percent of the
participants had been hospitalized during the first year
of at least one of their children. Many of the
participants (48%) had been hospitalized before their
first child was born, indicating that psychiatric
difficulty often preceded the onset of parenthood
status.
Current Parenting Context

Few of the participants were parenting their
children with the support of a spouse: only 21% of the
participants were married, and 79% were single,
separated, or divorced. Racial differences in marital
status were observed {X/3(3) = 16.5, p < .001} (See
Table 2). Among black participants 94% were single
parents as compared to 53% of the white participants.
Seventy-one percent of the black participants as
compared to 17.6% of the white participants had never
married. Among the participants who had ever married,
77.8% of the black mothers and 42.9% of the white
mothers were currently separated or divorced. (See
Table 2.)

Marital status was related to who cared for the
participants' children during the current

hospitalization {12(4) = 25.5, p < .001}. Relatives,

Ere

 

Hf:-

55

Table 2.

Frequencies of Marital Status by Race

MARITAL STATUS

Single Married Separated Divorced

 

 

 

 

 

 

 

 

RACE
22 2 3 4
Black
71.0% 6.5% 9.7% 12.9%
3 8 l 5
White
17.6% 47.1% 5.9% 29.4%
25 10 4 9
TOTAL

52.1% 20.8% 8.3% 18.8%

TOTAL

31

64.6%

17

35.4%

48

56

often the participant's mother, were relied on most
frequently (66.7%) for childcare. This was true
particularly for single mothers (96%), and to a lesser
extent for mothers who were no longer married (46.2%).
Fathers of the children provided childcare most
frequently for married mothers (70%). Foster care was
seldom relied on by the total sample (8.3%), but was
used most frequently by no longer married participants
(23%).

The majority of mothers were attempting to raise
their children in impoverished circumstances.
Sixty-seven percent of the mothers were receiving some
form of public assistance. Although 85% of the mothers
had been employed in the past, only 13% had been
employed immediatedly prior to the current
hospitalization. Those who were unemployed had been so,
on the average, for 5.8 years, and 58% of these
participants had been employed for less than a year at
their last job. Forty-six percent of the participants
reported that their financial resources were not
adequate to provide for the needs of their families.

Fifty-four percent of the participants rented or
owned their residence: the remainder lived with
relatives or friends. Thirty-five percent of the
participants lived with their own mothers, indicating

that they were raising their children in the context of

57

their family of origin. Most of the participants (74%)
felt their living situation had adequate space, and 65%
reported that they reported that in-town transportation
was easy (65%). Although 67% of the participants rated
their living situation as a good place for children to
grow up, only half (49%) reported the area to be a safe
place to live.
Social Support

The social support networks reported by the mothers
varied widely in size (fl_= 11.5: range = 2 - 38). The
participants' support networks were dominated by
relatives (66.3%). On the average, the participants
reported 7.6 kin and 3.9 nonkin supporters. Sixty-two
percent of the participants reported only 3 or fewer
nonkin supporters. The participant's mother was
included as a supporter by 64% of the participants,
whereas the participant's father was listed as a
supporter by 30% of the sample. The participant's
children were frequently included as providing support:
all but 23% of the mothers listed their children as
supporters. Thirty-two percent of the participants
included a professional as a supporter. Neighbors were
included as supporters by 47% of the participants: 81%
included individuals with whom they lived as providing
support. 0n the average, 20% of the supportive

individuals in the participants' networks were persons

58

with whom they lived.

Table 3 presents the correlations between different
measures of the types of social support provided. More
specifically, the support measures are total number of
supporters (NSUPP): the numbers of supporters providing
the five types of social support: emotional support
(NEMOT), childcare assistance (NCHA), companionship
(NCOMP), information and advice (NINFO), and tangible
aid (NAID): and the number of supporters who cause
difficulty and distress (NDIFF). Having more
individuals who provide support in one area was
generally related to having more supportive individuals
in other areas.

Forty-one percent of the supporters provided only
one type of support, whereas 59% provided multiple forms
of support. Overall, the percentages of supporters
providing each type of support were as follows:
emotional support, 50%: companionship, 46%: child
assistance, 34%: tangible aid, 34%, and information and
advice, 28%. Most of the mothers (70%) reported at
least one individual who made their lives difficult: 51%
of these stressful network members also were sources of
positive support.

Most of the network members (67%) were seen at
least weekly with 40% seen daily. Finally, the mothers

reported satisfaction with the majority of their

59

Table 3.

Intercorrelations of Types of Social Suppport

NEMOT .67

NCHA .79 .53

NCOMP .68 .65 .45

NINFO .79 .50 .79 .36

NAID .86 .59 .84 .62 .67

NDIFF .54 .08 .41 .28 .51 .51

NSUPP NEMOT NCHA NCOMP NINFO NAID

60

supporters (78%).
Psychological Distress and Self Esteem

The SCLlO provided an index of the severity of
psychological distress experienced by the mothers. The
mean score for patients on this scale was 14.9 (sd =
8.6) which was consistent with the results of the
standardization of the instrument on inpatient and
outpatient samples by Nguyen, Attkinson, and Stegner
(1983), (g =14.5, gg.= 9.2). Collectively, the
participants had a chronic history of hospitalization:
the average number of psychiatric hospitalizations was
6.8 (range: 1 - 24). Because the number of lifetime
hospitalizations is influenced by the participant's
current age, the number of psychiatric hospitalizations
in the last three years was ascertained. The mothers
had an average of 3.04 hospitalizations (sd =1.69) in
the last three years, indicating that their disorders
were an ongoing source of disorganization in their lives
and that they frequently were separated from their
children. The mean number of days hospitalized in the
last three years was 242 (£9 = 271.2).

The distribution of scores on the self esteem scale
was equivalent to the distribution of scores reported by
Rosenberg (1965). Fifty-four percent of the mothers
scored below the two highest levels of self esteem.

Self esteem and psychological distress (SCLlO) were

61

significantly correlated (r = -.53). However, this
relationship may be largely spurious, as both self
esteem and psychological distress were associated with
attachment-related constructs and perceptions of the
parenting the mothers received as children.

Tables 4 and 5 present hierarchical multiple
regressions for self esteem and psychological distress,
respectively. In both regressions, variables related to
the participant's own childhood were entered first,
followed by stepwise entry of the remaining variables
(SINGLE and NDIFF in the regression for SCLlO). The
perceptions of her own mother as providing lax
discipline (LDM) and negative control (NCM) were
negatively related to self esteem, whereas the
perception of being close to someone as a child was
positively related to self esteem. Conversely for
SCLlO, perception of negative control (NCM) from her own
mother was positively related to psychological distress,
whereas the perception of being close to someone as a
child was negatively related to psychological distress.
These attachment-related constructs accounted for 40%
and 42% of the variance in self esteem and psychological
distress, respectively. An additional 13% of the
variance in psychological distress was accounted for by
single marital status (SINGLE) and the number of network

members who were sources of distress (NDIFF).

1
J

1513

62

 

Table 4.
Multiple Regression to Predict Self Esteem
Step Variable E to 2 fl AR
Enter '—
l. LDM 9.62 .003 .42 .18
2. CLOSE 7.98 .007 .56 .13
3. NCM 6.06 .018 .63 .09
g? = .44
Adj. 32 = .38
F (3,42) = 9.15,

Final Step
,5 E
-.37 .006
.37 .006
-.30 .023
E. < .001

1‘

I).

l".

.51.

Table 5.

63

Multiple Regression to Predict Psychological Distress

(SCL-lO)
Step Variable E to
Enter
1. NCM 13.52
2. CLOSE 11.18
3. SINGLE 5.22
4. NDIFF 5.22

2. 55
.001 .50
.002 .65
.028 .70
.028 .74

3?

Adj. 52

_F_‘_ (4,37)

.25
.17
.07

.06

.55
.50

11.42,

Final Step
/3 P.
.33 .01

-.35 .003
.29 .019
.26 .028

p < .001

D)-
SM

“HP:
5.0

per

.“ I’
(D
(All

 

6L,

Perception of Child Behavior Problems

The-gfscores for the total child behavior problem
scale of the Child Behavior Checklist (CTOTT) were
anticipated to indicate the degree of stress the
participant might experience from a difficult child,
thereby negatively influencing parenting. However,
CTOTT was not related significantly to any of the
parenting measures. A multiple regression was conducted
to explore factors that would relate to the participant
perceiving difficulty in her child. (See Table 6). The
participant's perception of negative control from her
mother (NCM) and the number of social supporters
providing child care assistance (NCHA) were positively
related to higher levels of perceived difficulty in the
child, whereas the composite measure of adequacy or
satisfaction with aspects of the living situation
(LIVADEQ) was negatively related to perceived child
difficulty. Together, these factors accounted for 40%
of the variance in CTOTT.
Measures of Parenting

This section reports the results of analyses of the
various parenting measures administered to the
participants. Initially, the participants' scores on
the four MCRE subscales are examined. Subsequently, the
following parenting measures are compared: the two

derived subscales from the MCRE: Overprotection-

U0

65

Table 6.

Multiple Regression to Predict Perception of Behavioral

 

Difficulty in the Child (CTOTT)

Step Variable _F_ to p_ Q AR Final Step
Enter 3 p
l. NCM 9.02 .005 .46 .21 .27 .063
2. LIVADEQ 7.69 .009 .60 .15 -.41 .006
3. NCHA 5.23 .028 .69 .09 .31 .028
3? = .45
Adj. 5? = .40
F (3,32) = 8.72, B.‘ .001

66

Overcontrol (OVPOVC) and Hostile Objectification
(HOSOBJ), the Difficulty subscale of the Parent Attitude
Survey (PASD), the Sensitivity to Children physical
punishment index (STCPUN), and the Sensitivity to
Children parenting style index (STCSTYLE). Separate
multiple regression analyses are presented for the
parenting variables OVPOVC, HOSOBJ, PASD, and STCPUN.
STCSTYLE is analyzed separately with path analysis.
Despite the fact that the four MCRE subscales
(Acceptance, Overprotection, Overinvolvement, and
Rejection) were evaluated and found to have poor
psychometric properties, the MCRE is the one instrument
of the obtained parenting measures with norms that allow
a comparison of the participants with a normal sample.
Briefly, the mean grscores of the participants for the
four MCRE subscales were as follows: 41.4, 63.8, 62.2,
and 59.9, respectively for Acceptance, Overprotection,
Overinvolvement, and Rejection In comparison to a mean
g-score of 50 with standard deviation of 10, the
participants scored nearly a standard deviation below
the mean for Acceptance, and near or above one standard
deviation for the remaining three scales presumably
tapping aspects of rejection. These results suggest
that the participants as a group maintained
substantially problematic attitudes in regard to

parenting and the relationships they have with their

child
extre
of th
stand

sampl

£83511
expec
many

low m
tetwe
diffe
GVPOV
indic
and c
diffi
cont;
[Elat
30505
StYle
bY ac
StruC
H53 1

dim

67

children. However, these results must be regarded with
extreme caution because of the questionable properties
of the four MCRE subscales and the original
standardization which was based on a small, middle class
sample.

The correlations between the various parenting
measures are presented in Table 7. As would be
expected, the measures are not independent. However,
many of the correlations are either nonsignificant or of
low magnitude, and different patterns of relationships
between the variables are evidenced: this suggests that
different aspects of parenting are being assessed.
OVPOVC and PASD were strongly related (g = .58),
indicating that participants who maintain overprotective
and controlling attitudes were also experiencing
difficulty and uncertainty in their parenting role. In
contrast, HOSOBJ and PASD were not significantly
related. The correlation of -.58 between STCSTYLE and
HOSOBJ may provide some validation of the parenting
style measure. Authoritative parenting, characterized
by acceptance of the child while providing firm
structure and guidance (i.e., a high score on STCSTYLE)
was inversely related to attitudes of hostility,
distance, and annoyance (i.e., a high score on HOSOBJ).

In each of the regression analyses of parenting

measures reported below, variables from the different

U

;a.le
C‘I'PCV!
STCP"

E PbCFL‘.

68

Table 7.

Intercorrelations of Parenting Measures

OVPOVC 1.00

HOSOBJ .26 1.00

PASD .58 .22 1.00

STCPUN .30 .15 .18 1.00

STCSTYLE -.09 -.57 -.05 -.28 1.00
OVPOVC HOSOBJ PASD STCPUN STCSTYLE

.01
< .001

I'OI'OI'O
A

domai
exami
corre

selec

child
the c

prece

ever;
Table
cont:
9arti

iCCOL‘

69

domains hypothesized to influence parenting were
examined for possible inclusion, and those significantly
correlating with the dependent variable of interest were
selected for use in the analyses. In general, variables
representing the mother's experience in her own
childhood were entered first into the analyses based on
the conceptual premise that these features temporally
preceded other influences.

The results of the multiple regression for
overprotection - overcontrol (OVPOVC) are presented in
Table 8. Positive involvement (PIM) and negative
control (NCM) perceived to characterize the
participant's mother were entered first, and together
accounted for 26% of the variance in OVPOVC. These
seemingly divergent constructs both influenced OVPOVC in
a positive manner. The patient characteristics of race,
education, and diagnosis were subsequently added in a
stepwise entry procedure, and together they accounted
for an additional 18% of the variance. Education and
diagnosis showed trends toward significant
contributions. Minority status, less education, and
schizophrenic diagnosis were related to higher levels of
overprotective-controlling parenting attitudes. Two
other variables, current age and age at first birth,
correlated relatively strongly with OVPOVC (-.45 and

-.43, respectively). However, these variables did not

Table
hulti;

_

Overcc
‘

Step

Table 8.

70

Multiple Regression to Predict Overprotection -

 

Overcontrol (OVPOVC)

Step

Variable

PIM

NCM

RACE

EDUC

DX

.029
.006
.036
.082

.075

Adj.

.33
.51
.59
.62

.67

Iwm [SUN

5 (5,37)

.11
.15
.08
.05

.05

.44
.37

5.91,

.001

Final Step
£5 9
.20 .16
.21 .14

-.22 .10
-.25 .06
.26 .08

71

contribute to the regression analysis: their influences
were explained by their relationships to the other
variables.

Similarly, NCM, LDM, and CLOSE were significantly
correlated with HOSOBJ. Each of these variables related
to the participant's perceptions of her childhood and
her mother and were entered first into the regression of
HOSOBJ, as shown in Table 9. Together, these variables
accounted for 37% of the variance, or 82% of the total
a? of .45. Education contributed an additional 8% to
the variance accounted for. Thus, perceptions that the
participant's mother was lax in discipline and
negatively controlling and the absence of someone to
whom the patient felt close in childhood, as well as low
education, most powerfully predicted hostile parenting
attitudes in the participants. Two other variables,
psychiatric hospitalization while pregnant and self
esteem, were moderately related to HOSOBJ, correlating
.35 and -.35, respectively: however, they did not
contribute further to predicting the dependent variable.

The Difficulty scale of the Parent Attitude Survey
(PASD) was subjected to a multiple regression analysis
in order to examine influences on the participants'
perceptions of stress, difficulty, and uncertainty in
their parenting roles. The results of this analysis are

presented in Table 10. Two variables, negative control

Table 9.

72

Multiple Regression to Predict Hostile-Objectification

(HOSOBJ)

Step Variable ‘5 to
Enter

1. LDM 17.93

2. CLOSE 5.14

3. NCM 3.22

4. EDUC 5.78

2 53
.001 .44
.029 .56
.080 .61
.021 .67

32
Adj. 52
g (4,38)

.44

.39

7.76, 2'< .001

Final Step
,6 E
.35 .008

-.29 .021
.23 .070
-.30 .021

~\.U

Table 10.

73

Multiple Regression to Predict Parenting Difficulty

 

(PASD)
Step Variable _§ to
Enter
1. NCM 6.25
2. EDUC 10.74

p. 15
.016 .35
.002 .54
2
5
, 2
Ad]. 3
F (2,44)

2
A5 Final Step
73 P.
.12 .34 .011
.17 “.42 .002
.29
.26

9.17, p < .001

experi
(NCM)

regres
in p81
contr:
were '
other
did n
regre
aarri

fathe

Sigl

to“,

74

experienced from the participant's mother in childhood
(NCM) and education, significantly contributed to the
regression equation, accounting for 29% of the variance
in perceptions of parenting stress. Greater negative
control from the participant's mother and less education
were related to higher scores on PASD. A number of
other variables, although moderately related to PASD,
did not add further to the explained variance of the
regression equation (minority status, younger age, never
married, low SES, psychological distress, and loss of
father in childhood).

STCPUN provided an indication of the likelihood of
using physical punishment in disciplining children. As
with PASD, minority status, younger age, and never
having married were all related to higher scores on
STCPUN. As race, age, and single marital status were
intercorrelated significantly (5's ranged from .34 to
.51) and because young, single, black mothers comprised
29% of the sample, the summary variable of single,
black, and young (SBY) was created. The multiple
regression for STCPUN is summarized in Table 11. SBY
‘Mas selected first for this stepwise regression. Public
assistance and social support from her children also
entered into the equation but were of marginal
Significance. Together, these three variables

contributed to an R? of .39. Single, black, and young

Table
hultig

15181:

~

Step

Table 11.

75

Multiple Regression to Predict Physical Punishment

(STCPUN)
Step Variable §_to
Enter
1. SBY 10.09
2. PUBAST 3.66
3. SSCHLD 4.09

a ES
.003 .48
.065 .56
.052 .63

52
, 2

Ad3.,§

g (3.31)

.23
.08

.08

.39
.33

6069’

IU

Final Step
/.’o P.
.25 .137
.31 .043

-.31 .052

< .001

inv
inc
ine
foc
sec
sel
se;
his
Psy
Slip
inf
dom
sel

has

76

as well as receiving public assistance was associated
with greater likelihood of use of physical punishment,
whereas social support from the mother's children was
negatively related to physical punishment.

Path Modeling of STC Parenting Style.

The composite measure of STC parenting style
(STCSTYLE) was chosen as the parenting measure to use in
investigating the proposed path model because it
incorporates multiple components of effective vs.
ineffective parenting into one measure rather than
focusing on a single aspect of parenting. (See Method
section for a description of STCSTYLE.) Variables were
selected for the causal models in order to represent the
separate domains influencing parenting {developmental
history, personality adjustment (degree of
psychopathology, self esteem), social context (social
support, physical living environment), and child
influences} as proposed earlier. In some of the
domains, multiple variables were available for
selection. Variables were chosen to represent a domain
based on two considerations: the generality of the
content implied by the variable (rather than specificity
bolone discrete feature of the domain) and evidence of
cxorrelation with either STCSTYLE or other variables
being considered for the model. Table 12 presents a

list of potential variables in each of the domains

Table l

Potenti

Ill
NCM

LDM
CLMOM
CLDAD
CLOTHR
CLOSE
HARHAR
330M
QUAD
PSYHOM

Table 12.

77

Potential Variables for Path Models for Each Domain

Develomental History

 

PIM
NCM
LDM
CLMOM
CLDAD
CLOTHR
CLOSE
MARHAR
QMOM
QDAD
PSYMOM
PSYDAD
MSEPS
MSEPlZ
FSEPS
FSEPlZ
SEPBPS

SEPBFlZ

MLOSS
FLOSS
LPRNTS
MARDEV
FCEV

*

Positive involvement with mother as a child
Negative control from mother as a child

Lax discipline from mother as a child

Close to mother as a child

Close to father as a child

Close to other person as a child

Close to anyone as a child

Rating of parents' marital harmony as a child
Rating of quality of relationship with mother
Rating of quality of relationship with father
Psychiatric history in the mother

Psychiatric history in the father

Separated from mother under age 6

Separated from mother under age 13

Separated from father under age 6

Separated from father under age 13

Separated from both parents under age 6
Separated from both parents under age 13

Loss of mother in childhood

Loss of father in childhood

Lived with both parents while growing up
Parental marital disruption in childhood

Ever placed in foster care

Psychopathological Status

AGEHOSPl

YRSSNC
LLONG
NHOSP
NADM3
DAYH3
SCLlO

Age at first psychiatric hospitalization
Years since first hospitalization

Lenght of longest hospitalization

Number of hospitalizations

* Number of hospitalizations in last 3 years

Days hospitalized in last 3 years
Total score from SCLlO

78

Table 12 (cont'd.)

Social Support

NSUPP
NKIN

NNKIN
SSMOM
SSDAD

SSSPOUS

SSBF
SSCHLD
NPROF
NCOMP
NINFO
NAID
NCHA

NEMOT
NDIFF

NHISAT
NLOSAT

NWEEKLY

Total number of suppporters in network

Number of kin in network

Number of nonkin in network

Social support from mother

Social support from father

Social support from spouse

Social support from boyfriend

Social support from child(ren)

Number of professionals in network

Number of supporters providing companionship

Number of supporters providing information

Number of supporters providing aid

Number of supporters providing childcare
assistance

Number of supporters providing emotional
support

Number of stressful network members

Number of supporters satisfied with

Number of supporters unsatisfied with

Number of supporters contacted weekly

Child Behavioral Problems

CINTT
CEXTT
CTOTT
CSOCCT

T-score of internalizing problems (CBCL)
T-score of externalizing problems (CBCL)

* T-score of total behavior problems (CBCL)

T-score of social competence (CBCL)

Physical Living Environment

PUBAST
MONEY

SPACE

TRAVEL
LIVSAT
GROWUP
SAFETY

EASCARE
LIVADEQ *

NLIV
PLLIV

LIVSTAT

Note.

Receiving public assistance

Adequacy of financial resources
Adequacy of space at home

Ease of in-town transportation
Satisfaction with living situation
Quality of home for child to grow up in
Safety of neighborhood

Ease in taking care of children

Summary measure of adequacy of liv. situation
Number of persons living in residence
Type of household living in

Renting, own, live with relatives

Starred variables were selected for inclusion in

path models.

COF

V3!

to

was

Ste

38

79

considered for inclusion in the model with the selected
variables indicated.

Variables chosen to represent experiences from the
mother's childhood included perceived maternal lax
discipline (LDM), a history of significant separation
from the participant's mother occuring through age 12
(MSEPlZ), and the patient's perception of having someone
to whom she was close as a child (CLOSE). Self esteem
was measured by the scale score from the Self Esteem
Scale (SLFESTM). The number of psychiatric admissions
in the past three years (NADM3) was used as an index of
the severity of the patient's disorder. Finally, the
number of persons in the mother's social support network
who provided emotional support (NEMOT) was selected to
represent the extent of global social support available
to the mother.

Three aspects of the proposed model SES, perceived
adequacy of the physical living environment (LIVADEQ),
and perceived child behavioral difficulty (CTOTT) were
not significantly correlated with STCSTYLE (E's = .12,
-.12, and .04, respectively). Further, these variables
did not evidence relationships with the other variables
to be included in the model: 17 out of 18 of the
intercorrelations were not significant. As a result,
these components of the proposed model were excluded

from further analysis. Race also was not related

80

significantly to STCSTYLE (g = .11, n.s.).

For the final set of variables selected, MSEPlZ,
CLOSE, LDM, NADM3, SLFESTM, NEMOT, and STCSTYLE,
intercorrelations are presented in Table 13. When using
path analysis to model causal relationships, it is
advisable to correct the correlations for unreliabilty
of measurement so that the relationships between
components of the model are based on true scores rather
than observed scores. (Observed scores contain
measurement error and, therefore, create a systematic
downward bias.) Reliability estimates were available
for scale scores used in the model, LDM, SLFESTM, and
STCSTYLE (Alpha = .67, .62, and .61, respectively). It
was not possible to assess reliability for the other
measures in the model and therefore, their reliabilities
were assumed to be 1.0. Correction for attenuation in
the correlations is obtained by dividing the observed
correlation by the product of the reliabilities of the
two measures being correlated (Ghiselli, Campbell, &
Zedeck, 1981). The corrected correlations among
variables in the analysis also are presented in Table
13. The path models to be examined will be based on
these corrected correlations.

Investigation of the causal patterning among the
variables proceeded in two steps. First, proximal

components (SLFESTM, NEMOT, NADM3) influencing parenting

81

.Hmcommfin ecu co mum moumeumo
>uwuwnmd~mu «accomec ecu m>onm mum cofiumsucmuum LON cwuumuuou
mcofiumHmuuoo «Hmcommfio mcu zoaoo mum mc0wu0~muuoo po>uomno .wuoz

mawamosm eozmz sanqum mza<z so; mmoao madam:

146.2 me. mm. md.: 04.: ms. 04.: mssamuam
mm. 206.43 ov. 64.: m~.u am. vm.u eozmz
mo. Hm. Amo.v mo. ~v.u He.u no.1 zammmdm
mm.u v~.e mo. 266.42 oH.n oH.u do. mzoaz
~0.u m~.a 66.: «4.: 256.2 ~4.u 43.: and
dm. mm. mm. oa.u mH.e 260.42 m~.n mmoao
mH.u vm.n 66.: co. va.e ma.e loo.av mammmz

 

mampoz cums ecu cw 00H90wum> no mcoquHouuooumumw

.ma manna

82

were examined alone to investigate how these current
features in the participant's life predicted parenting.
Second, the distal variables related to experiences in
the mother's childhood (MSEPlZ, CLOSE, LDM) were added
to the proximal variables to examine how these
historical influences related to parenting and the
proximal predictors of parenting.

Three causal patterns among the proximal variables
were examined. Figure 4 presents the three path
analyses conducted by the ordinary least squares method.
In all of the models, each proximal variable has a
direct influence on STCSTYLE. Each model ascribes
primacy in the causal patterning to a different proximal
variable. In Model A, SLFESTM is the exogenous
variable, directly influencing NEMOT and indirectly
influencing NADM3. Through these influences on NEMOT
and NADM3, SLFESTM also indirectly influences STCSTYLE.
In contrast, Model B places NEMOT as the exogenous
variable directly influencing both SLFESTM and NADM3
with consequent indirect effects on STCSTYLE. Finally,
in Model C, NADM3 takes the position of the exogenous
variable with a direct effect on NEMOT and an indirect
effect on SLFESTM. Other possible causal orderings
among the predictor variables were examined but were
found not to reproduce the original correlations as well

as the three models under consideration.

83
Figure 4. Path Models of Proximal Variables

MODEL A

SLFESTM

MODEL B

 

MODEL C

 

Note. Figures in parentheses are path coefficients based on
uncorrected correlations.

+2<100 *2<005o “2(001

Pi

all

ea
In
to
an
Th
di.
Vhe
Pre
leg
91a

[69

84

The path coefficients in the path diagrams of
Figure 4 indicate the strength of direct influences
among the variables. The path coefficients represent
beta weights for regressions of the endogenous variables
in the analysis. As it is not feasible to perform
regressions with the correlations corrected for
attenuation, the significance of the path coefficients
represents the significance of the betas for regressions
conducted with the observed data. If the betas obtained
with observed data are significant, it is assumed that
the path coefficients based on the corrected
correlations are significant.

Despite the different causal directions of
influence among the variables posited by the three
models, the values of the path coefficients remain the
same for each path between variables across the models.
In each of the models, self esteem is positively related
to parenting style, whereas NADM3 is negatively related,
and NEMOT has a weaker yet positive effect on parenting.
Thus, higher self esteem and more emotional support
directly contribute to effectiveness in parenting,
whereas more frequent psychiatric admissions, and
presumably greater severity of disorder, contribute to
less effective parenting. Each of the models is equally
plausible in explaining the data as each model

reproduces the original correlations between the

varia
model
the 0
each

relat
resul
posit

NEMOT

is st
as ir
the a
repre
on pa

Other

°rigii
depenr
°rigir
Spuric
the Qt
Variak
prESen
additis
inflUer

n° indj

85

variables with the same accuracy: the index of path
model accuracy, the sum of squared deviations between
the original and the reproduced correlations, is .01 for
each of the models. In all three models, a strong
relationship is found between SLFESTM and NEMOT, and the
results are not altered depending on whether SLFESTM is
posited to influence NEMOT as in Model A, or whether
NEMOT is posited to predict SLFESTM as in Models B and
C. Although the relationship between SLFESTM and NEMOT
is strong, these variables are weakly related to NADM3
as indicated by the nonsignificant path coefficents in
the models for paths linking NADM3. Thus, NADM3
represents a relatively independent source of influence
on parenting, and does not influence substantially the
other independent variables in the models.

With path analysis, it is possible to decompose the
original correlations between the independent and
dependent variables into the degree to which the
original relationship is a result of direct, indirect,
spurious, or unanalyzable effects. The decomposition of
the original correlations between the independent
variables and STCSTYLE for the separate models is
presented in Table 14. In Model A, SLFESTM has
additional indirect effects on STCSTYLE through its
influence on NEMOT and NADM3, whereas in Models B and C,

no indirect effects exist and the remaining degree of

86

Table 14.
Decomposition of Relationships between the Independent

Variables and Parenting Style (STCSTYLE)

SUM OF SUM OF
DIRECT INDIRECT SPURIOUS REPRODUCED
EFFECTS EFFECTS EFFECTS 5
SLFESTM to
STCSTYLE
Model A .52 .15 .00 .67
Model B .52 .00 .15 .67
Model C .52 .00 .15 .67
NEMOT to
STCSTYLE
Model A .21 .07 .27 .55
Model B .21 .34 .00 .55
Model C .21 .27 .07 .55
NADM3 to
STCSTYLE

Model C -.53 -.07 .00 -.60

87

influence between SLFESTM and STCSTYLE is spurious,
resulting from SLFESTM's relationship with NEMOT and
NADM3. Similarly, for NEMOT the models differ by the
degree to which the original relationship is a result of
indirect or spurious influences of NEMOT on STCSTYLE.
Model B suggests the strongest indirect effects of NEMOT
on STCSTYLE whereas in contrast, Model A suggests the
majority of NEMOT's influence is spurious. Finally,
only a small degree of indirect or spurious influence
(-.07) is found for NADM3 regardless of the model under
consideration. In summary, it appears that the greatest
difference between the models occurs in the
decomposition of effects for NEMOT, i.e., the degree to
which NEMOT is related to STCSTYLE in an indirect or
spurious manner.

The complete models are constructed by linking the
distal variables related to the participants' childhood
experiences (MSEP12, CLOSE, LDM) to the above proximal
models, A, B, and C. These three new models are labeled
Models A2, B2, and CZ and are presented in Figures 5, 6,
and 7. For comparison purposes, the same paths from the
distal variables to the variables in the original
proximal models are included across the three new
models. The paths linking the distal variables to the
variables in the proximal models include the following:

(1) separate paths from LDM to SLFESTM, NEMOT, and

88

 

éfivm+

 

Hoo.V A .12.. .Ho. V Al: .no.V we:
.mcogmaosnoo 60.60.5005. :0 0003 0233.380 589 and @33ch 5 mafia .302

:5

 

 

«amiss...

 

:32.

(av-lev-

(II'-‘)v1'-

 

(81")81"

8 .-
* Ma QM: .1
«ram:

.~< dado: seem .n magmas

89

 

.«oo.VN*** .«o.VN** .no.vm.* .o«.VN+
.mcowadaonuoo cowoouuooc: :0 comes mecoaoammooo spam one monocacmadm ca megawah .oaoz

 

 

LINM.-Vsm.-

A223. 595m“ .

 

 

 

(II'-)vt'-

Nahum:

 

mm 88: 58 .6 858E

90

 

 

:25. v N I; .3. v m. 1:. .no.v N * .o«.V N +
.mcogdaouhoo 63008825 :0 comes mvcoaofiwmooo 5.8m 0H0 monocpcofia 5" madman .302

 

 

ANM.-Vsm.-
x.

 

Aka.sma. sandman

.. 3.32%.- .
«gamma

No Adda: sham .R enemas

(21°?)91'-

(II'-)v1'-

(81'-)8I°-

 

91

STCSTYLE: (2) separate paths from CLOSE to SLFESTM and
NEMOT: and (3) a path linking MSEPlZ to NEMOT. Note
that the distal variables, related to childhood
experiences, are all considered exogenous variables: no
causal pattern of influence is posited between these
variables which are all weakly and negatively related.
As previously, the three models differ by the posited
direction of effects between SLFESTM, NEMOT, and NADM3.

The three path models to be discussed are not the
only possible models that exist between the variables.
The goal of path analysis is to attain a parsimonious,
plausible description of directional influences among
the variables. If paths between all variables are
posited (a just-identified model), there is no way to
evaluate the effectiveness of the model because
just-identified models will always reproduce the
original correlations between the variables perfectly
(Pedhazur, 1982). Over-identified models eliminate some
of the possible paths based on theoretical grounds and
are valuable in that they simplify the explanation of
the phenomenon under condsideration. Further, they can
be tested for goodness of fit, the degree to which the
posited relationships reproduce the original
correlations among the variables.

With the current set of variables, other possible

models were examined which included both additional and

92

fewer paths between the variables. Based on theoretical
and conceptual grounds, some paths were not included.
For example, if a path is postited from LDM to NADM3,
the overall accuracy in each of the three new models was
improved (sum of squared deviations = .02, .01, and .03
for Models A2, 82, and 02, respectively). However, the
conceptual rationale for this link is weak and, thus,
the path was excluded. Further, a link between MSEP12
and SLFESTM is theoretically possible, but the direct
path, when included, was found to be weak (i.e., path
coefficient in each of the models < 1.10) and, thus, was
eliminated. Similarly, direct paths could be posited
from both MSEP12 and CLOSE to STCSTYLE. Again, however,
the coefficients of these paths were very weak, and the
paths were eliminated. Thus, the three new models to be
explored retained important conceptual links while also
eliminating paths that were found to be very weak
statistically.

The additions of MSEP12, CLOSE, and LDM in the
three causal models influenced the strength of
relationships between STCSTYLE and SLFESTM, NEMOT, and
NADM3. Although only minor fluctuations in the size of
the path coefficients from NEMOT and NADM3 were noted,
the path from SLFESTM to STCSTYLE was reduced
substantially (.52 vs. .18). This change resulted from

strong direct links between LDM and SLFESTM and LDM and

93

STCSTYLE, relegating much of the original relationship
between SLFESTM and STCSTYLE to a spurious source. The
three new models also evidenced variation in the size of
the path coefficients between SLFESTM and NEMOT
depending on the direction of causal influence. The
magnitude of the coefficient was higher in Model A2
(.41) with SLFESTM predicting NEMOT than it was in
Models 82 and 02 (.22) with NEMOT predicting SLFESTM.
This difference also was related to variation between
Model A2 and Models BZ and C2 in the strength of
influence of LDM on NEMOT and CLOSE on NEMOT: for both
paths, the coefficients in Model A2 (-.04 and .12) were
weaker than in Model BZ (-.28, .29) and Model C2 (-.30,
.27).

The models each reproduced the original
correlations almost equally well and within acceptable
limits: the sum of squared deviations was .07, .07, and
.05, for Models A2, 82, and CZ, respectively. The
decomposition of the correlations of STCSTYLE with the
other variables in the path models are presented in
Table 15. By comparing the size of direct, indirect,
and spurious and unanalyzable effects for the various
variables, it can be seen that the three models were in
large part very similar. The most notable variation
between the models occurs for the variables SLFESTM and

NEMOT in the degree to which their original

94

no. mo.l 00. mo. HH.I OO. 50.
NH. vH. 00. NH. 5H. 00. NH.
00. wH.I Hm.l OO. ON.I Hm.l HO.I
OO. vo.l mm.l 50.! OO. mm.l 50.!
SN. v0. VN. mH. NH. vN. MN.
5%. OO. mH. av. 00. mH. mm.
dbam mH mo <Dwm mH mo flowm
No ammo: Nm ammo:

OH.I

MH

N4 AMQOZ

OO. NHmmmZ
OO. mmOQU
Hm.l 20d
mm.l mzo<z
vN. BOZMZ
mH. 29mmmamm
mm

 

NO was .Nm .N< mH0002 numm 0:u no“ A4: a my muomuwm

 

oHnmnaHmcmcD 0cm msoHusmm 6cm .AMHV muomuum uoouHOCH .Amov muowuum uomuHa

 

OucH mqwamoem cuH3 moHomHum> LODOHcoum no mchmcoHumHom no coHuHmomeoooQ

.mH anma

95

relationships with STCSTYLE are accounted for by
indirect vs. spurious and unanalyzable effects.

Overall, the models indicate that STCSTYLE is
strongly and negatively influenced by both LDM and
NADM3, and weaker positive influences are exerted by
NEMOT and SLFESTM. CLOSE, LDM, and MSEP12 directly
influence current SLFESTM and NEMOT and thereby exert
weak indirect effects on STCSTYLE. The direct influence
of SLFESTM on STCSTYLE is reduced substantially in the
distal models as compared to the previous proximal
models. Although NADM3 strongly influences STCSTYLE, it

is relatively independent of the other variables.

DISCUSSION

Evaluation of the Hypothesized Model of Parenting

In studying the determinants of parenting in normal
parents, one is confronted with a plethora of potential
contributions to individual differences: the problem is
even more complex when determinants of parenting in
severely disordered mothers are considered. Belsky's
(1984) conceptual model provided a useful framework to
organize the diversity of influences explored in this
investigation. Rather than examining individual
influences in isolation, the use of path analysis in
this study allowed for an exploration of the mutual
interrelationships among the variables potentially
predictive of parenting style (maternal childhood
history, personality, social support, environmental
context, and child behavioral difficulty). This
strategy provided a more detailed understanding of
processes involved in promoting variation in parenting
effectiveness among disordered mothers. In general, an
altered version of the originally conceptalized model
was found to fit the data.

Initially, two pathways in the parenting model were
largely unrelated to the parenting style measure and

were excluded. First, socioeconomic status was

96

97

hypothesized to have a direct impact on the physical
living environment and, in turn, an indirect influence
on parenting style (STCSTYLE). However, the
relationships between socioeconomic status, a composite
measure of living situation adequacy, and parenting
style were found to be negligible. Because
socioeconomic status had a very restricted range (most
mothers were of low SES), it was unlikely that this
variable could demonstrate an influence on parenting.
Limited variation in socioeconomic status may have also
contributed to the negligible direct effect of living
situation adequacy on parenting. The variable of living
situation adequacy was composed of the mother's
perceptions of a number of features of her living
environment (e.g., financial resources, neighborhood
safety, satisfaction with living situation) that were
presumed to reflect potential sources of stress in the
mother's life. However, satisfaction with these general
environmental features was unrelated to parenting style
as well as the other parental attitude measures. Beyond
the mother's perceptions, the actual qualitative
features of the living environment are not known. Given
the low socioeconomic status of the mothers, supportive
physical resources may be scarce and a source of stress,
even though some mothers may not perceive the lack of

resources to be a problem or may have adapted to the

98

resources available. A comparison of the mother's
perceptions with an external evaluation of the living
environment would help to clarify these distinctions.
Further, a sample with a greater variation in
socioeconomic status might exhibit greater diversity in
the adequacy of the physical living situation:
influences on parenting might then become apparent.

The second pathway unrelated to parenting was the
effect of behavioral problems in the mother's child, as
measured by the mother's perceptions on the Child
Behavior Checklist. Only three of the mothers indicated
a degree of disturbance in the child within the clinical
range (i.e., T—score > 70). Overall, the mothers did
not report adjustment difficulties in their children
beyond that found in the general p0pulation. As with
adequacy of the living environment, the obtained measure
for child behavior problems was based on the mother's
perceptions, and an independent evaluation of the child
would help to determine the veridicality of her
perceptions. Nevertheless, the potential of stress from
perceived child behavioral difficulty (or lack of
difficulty) was not related to more accepting vs.
neglecting parental attitudes or to the other parenting
measures.

Although a range of differences were found for both

perceived child difficulty and adequacy of the living

99

environment, the majority of mothers did not acknowledge
difficulty in these areas. As Cohler et al. (1980)
reported, the disordered mothers may deny the presence
of problems and difficulty when, in fact, it may exist.
This further highlights the need for external evaluation
of the mothers' children and the mothers' home
environments. Lack of awareness or denial of problems
in these areas could potentiate increased difficulties
in parenting as a result of minor problems increasing in
severity if left unremedied.

Alternatively, the lack of demonstrated
relationships for child behavioral difficulty and
adequacy of the living environment may, in fact, reflect
their limited causal impact on parenting. Belsky (1984)
theorized that parental personality adjustment would be
of primary importance for competent parenting, followed
by social-contextual contributions, and finally, child
influences. Thus, the exclusion of the child behavioral
difficulty and adequacy of the living environment
components from the model is in keeping with Belsky's
predictions of their less powerful causal significance.
What remains in the model are, in fact, features of
personality adjustment and other aspects of the social
context (i.e., social support), as well as maternal

developmental history.

100

Proximal Models of Parenting

The proximal models of parenting (Models A, B, and
C) examined self esteem, emotional support, and number
of admissions in the last three years as predictors of
parenting style (see page83). The causal orderings
between these three components varied depending on
hypothesized directions of influence. In Model A,
higher self esteem was posited as promoting a greater
availabilty of persons in whom one could confide, and
this availability would buffer the psychological status
of the individual, preventing multiple hospital
readmissions. In contrast, in Model B, emotional
support was conceptualized as protecting the individual
from rehospitalization and contributing to greater
feelings of self acceptance and worth. Finally, Model C
posited that more frequent admissions would erode the
number of persons available to provide emotional support
leading to diminished self esteem. Each of these causal
orderings fit the data equally well. Rather than one
unidirectional pattern accounting for the causal flow
most accurately, cyclic relationships between the
variables are more likely to operate. It seems
plausible, for example, that having more persons
providing emotional support would contribute to greater
self esteem which, in turn, would lead to greater

receptiveness to caring from others, etc. From the

101

models, the strongest reciprocal influences are likely
to exist between self esteem and emotional support.
However, the strength of paths between emotional support
and number of admissions was weak in all of the models
regardless of the directional flow. This suggests that
continual rehospitalization is a relatively independent
phenomenon that strongly influences parenting but has
little impact on emotional support: conversely,
emotional support may provide little buffering effect on
rehospitalization. These tentative conclusions would
need to be substantiated with longitudinal research to
address directly potential cyclic influences among the
variables.

In all three of the proximal models, self esteem
and number of admissions in the last three years were
more strongly related to parenting style than was the
number of emotional suppporters. This finding again
echoes Belsky's (1984) predicitions of the primacy of
personality adjustment factors in determining parenting
effectiveness. Although the direct influence of
emotional support was weak, emotional support indirectly
affected parenting style via its influence on self
esteem and/or number of admissions, depending on the
model examined. In models B and C, the sum of indirect
effects was stronger than the direct effect. This

indicates that although mothers are more sensitive to

102

their children in the context of emotional support from
others, the influence of emotional support on parenting
also occurs indirectly by enhancing self esteem and/or

preventing hospitalization (albeit weakly).

In contrast, the direct effect of frequent
admissions in the last three years on parenting was
substantially negative. Indirect effects were
nonexistent or negligible. Frequent psychiatric
admissions as used here was considered an index of
severity and chronicity of the mother's mental disorder.
The direct relationship suggests that the more
instability in the mother's mental condition, the less
available she is to respond sensitively to her child.
Repeated separations as a result of hospitalization also
are likely to undermine the continuity of the
parent-child attachment relationship, perhaps provoking
uncertainty and guilt on the mother's part and
contributing to neglecting attitudes toward her child.

Self esteem, the other personality adjustment
component, directly affected parenting style positively:
small additional indirect effects were seen only in
Model A, in which self esteem positively influenced
emotional support and, in turn, affected parenting.
Thus, among this group of disordered mothers, those who
maintained a sense of self worth and self acceptance,

despite the debilitating effects of their disorders,

:2...

103

were more likely to respond to their children with
acceptance and sensitivity while also providing
structure for the child. Conversely, those mothers with
low self esteem were more likely to evidence avoidance
of their children and to be emotionally distant or
unresponsive.

The proximal path models examined how influences
within the recent temporal context were interrelated and
predictive of differences in parenting style. Aside
from the apparent bidirectional influences between self
esteem and social support, the proximal models do not
provide an understanding of sources of variation in self
esteem and emotional support. Further, as Scarr (1985)
has suggested, distal variables outside of the immediate
context may strongly influence current behavior and
diminish the relative strength of effects of more
proximal determinants. These potential distal
influences were incorporated in Models A2, B2, and C2 in
the form of variables related to the mother's childhood
experiences of significant relationships.

Distal Models of Parenting

The features of the mother's childhood included
three components. The first was a history of
significant separation from her mother occurring through
age twelve. Included here were also cases in which the

mother had lost her own mother as a result of death or

104

abandonment. Second, the mother's perception of having
a close relationship with someone as a child was
evaluated. This relationship did not have to involve
the mother's mother, and in the majority of cases it did
not. The issue of interest was whether there was a
close attachment bond with someone, perhaps to
compensate for a difficult or nonexistent relationship
with the mother. Third, the individual's perception of
her mother providing lax discipline was examined.
Although this variable ostensibly measured lax
discipline, it also may be thought of as an indication
of detachment or uninvolvement in the parenting
experienced.

In regard to the proximal variables, the major
impact of incorporating the historical influences was to
reduce substantially the direct effect of self esteem on
parenting style. This decrease was evidenced in each of
the models and resulted from the strong relationships
evidenced between lax discipline and self esteem and
between lax discipline and parenting style. In other
words, much of the influence of self esteem on parenting
style can be regarded as spurious because lax discipline
strongly influences both self esteem and parenting
style. Thus, the independent influence of self esteem
in the distal models dissipates because it is so highly

determined by the experience of the quality of

105

relationship with the mother in childhood. In each of
the models, self esteem was directly influenced by lax
discipline and a close relationship in childhood. A
close, ”attachment-like" relationship in childhood
contributed to higher self esteem whereas lax discipline
from the mother, and the detachment implied by such
parenting, contributed to lower self esteem. The
experience of separation from the mother was not found
to relate to self esteem.

Although not anticipated, the childhood relational
experiences predicted current emotional support from
others. In each of the models, childhood separation
from the mother was directly related to fewer emotional
supporters in the current network. Further, a close
childhood relationship and maternal lax discipline also
were directly related to emotional support: the strength
of the relationships varied depending on the specific
model examined, but the directions of effects were
consistent. Generally, a close relationship in
childhood contributed to greater current emotional
support, whereas maternal lax discipline or
uninvolvement contributed to fewer current emotional
supporters. These findings help to place emotional
support in a developmental context rather than viewing
emotional support as an environmental provision that is

more or less a result of one's social context. A mother

106

may be constrained from obtaining needed emotional
support to assist her in effective parenting as a
residual consequence of childhood experiences. These
findings are consistent with recent theoretical
discussions on social support (Sarason, Shearin, Pierce,
& Sarason, in press). These authors have emphasized the
central role of emotional support in the various
conceptualizations of social support, and they have
speculated that emotional support may represent adult
analogues of previous developmental attachment
relationships.

As with the proximal models, the distal models
differed only in the posited direction of effects
between self esteem and emotional support and number of
admissions. However, with the distal models the
strength of some of the path coefficients varied.
Specifically, the major difference pertained to the
direction of influence between self esteem and emotional
support posited by the different models: the
relationship was stronger with self esteem predicting
emotional support than the converse. This difference
also related to the strength of the direct effects of
maternal lax discipline and a close childhood
relationship on emotional support. The direct effects
of these variables were much weaker if self esteem

predicted emotional support. In this case, these

10?

variables evidenced greater indirect effects on
emotional support through their strong relationships
with self esteem.

Overall, the strongest direct effects on parenting
style resulted from the number of hospitalizations in
the last three years and maternal lax discipline
experienced in childhood. Emotional support provided an
additional but weaker direct influence. Self esteem's
strong influence in the proximal models was absorbed by
maternal lax discipline and was substantially
diminished. The other two variables, childhood
separation from mother and a close childhood
relationship, did not affect parenting style directly,
but weakly influenced parenting style indirectly through
their effects on self esteem and emotional support.

The inclusion of the distal variables related to
the mother's childhood added considerably to the
understanding of the processes influencing parenting
style among the disordered mothers and in a manner
consistent with Scarr's (1985) conceptualizations. When
the direct influence of child difficulty and adequacy of
the living environment were eliminated from the model,
the examination of the determinants of parenting shifted
to the greater influence of personality adjustment,
which in turn led to incorporating relational components

from the mother's childhood. This progression is in

108

keeping with Belsky's (1984) hierarchy of importance,
but greater emphasis has been found in the present
investigation for the influence of childhood
antecedents. In general, two basic lines of influence
emerged in the models. First, the number of
hospitalizations in the last three years remained a
virtually distinct component across the path analyses
and thus appears to operate independently. The second
line of influence incorporated the remainder of the
model: the childhood variables that influenced self
esteem and emotional support and which, in turn, were
related to parenting style. This line of influence
appears to be consistent with a pattern that may operate
for mothers in general, whereas the first line of
influence, to the degree that it represents severity and
chronicity of disorder, is an additional influence on
parenting for disordered mothers.
Determinants of Parenting: An Attachment Perspective
The overall results of the path analyses are
consistent with findings from a number of recent
investigations of normal mothers (Ricks, 1985: Main,
Kaplan, & Cassidy, 1985: Main & Goldwyn, 1984: Morris,
1982) that have explored continuities in infant - mother
attachment relationships and the mother's childhood
relationship with her own parents. These studies have

begun to explore a view of attachment relationships that

109

goes beyond the mother-infant dyad and includes parent
relationships with older children and attachment issues
in adulthood (see also, Heard & Lake, 1986). Although
the present investigation did not explore attachment
relationships per se, the conceptual dimensions inherent
in the parenting style measure are analogous to
differences found in the behavior of mothers in
attachment relationships labelled as secure vs. insecure
(anxious-avoidant or anxious-resistant). The above
investigations have demonstrated that mothers with
insecure attachment relationships with their children
evidence histories of problematic or rejecting
relationships with their own mothers or that their
recollections of childhood attachments are marked by
incoherence, distortions, or overidealization. This
suggests that they have difficulties integrating
contradictory affects and information about their
childhood attachment experiences. The authors contend
that internal representational models of relationships
are reenacted in later analogous situations,
particularly in the parenting role. Similarly, in the
present investigation, disrupted or absent attachment
relationships or attachment relationships with their own
mothers characterized by uninvolvement and detachment
were related directly or indirectly to neglecting and

unresponsive approaches to parenting.

110

Ricks (1985), in integrating both aspects of
attachment theory and self theory, found that not only
did mothers who had insecure attachment relationships
with their children evidence problems in their childhood
relationships with their own mothers, but also that
these subjects had lower self esteem than mothers who
had secure attachments with their children. Similarly,
lower self esteem in the present investigation was
related to more neglecting and unresponsive approaches
to parenting and problematic attachment relationships to
the mother in childhood. Consistent with attachment
theory, Ricks, in discussing internal representations of
relationships, predicted that difficulties in early
attachment relationships portend difficulty in later
adult-adult relationships. The present findings
indicated that mothers who evidenced difficulties in
their childhood attachments had fewer persons in their
current lives on whom they could rely for emotional
support. This supports the contention that internal
models of relationships originating in childhood
continue to influence later relationships, both with
other adults and with one's children. Processes that
contribute to less Optimal parenting relationships in
normal mothers are similarly related to differences in
parental functioning in mothers with severe

psychopathology. Interestingly, among the disordered

111

mothers in the current study, high levels of
psychological distress were predicted by difficulties in
their childhood attachment relationships (negative
control from their mothers, absence of a close childhood
relationship) and in their current relationships (more
stressful relationships in their social networks, never
having married).
Qgtachment in Childhood and Other Parenting Measures
Constructs related to the mother's attachment
experiences in her own childhood also predicted other
maternal attitudes toward parenting and relationships
with her children, as indicated by the regressions of
Hostile-Objectification, Overprotection-Overcontrol, and
Parenting Difficulty. These measures were postively
intercorrelated. Hostile-Objectification involved
endorsement of attitudes that signified rejecting,
negativistic attitudes toward the child and little
acknowledgment of the child as a unique, feeling person.
This measure was strongly correlated with the parenting
style measure used in the path analyses and was
predicted by some of the same variables, namely, lax
discipline experienced from the mother's own mother and
an absence of a close relationship in childhood. In
addition, parenting from the subject's mother
characterized by negative control also contributed to

Hostile-Objectification. Thus, uninvolvement,

112

detachment, and punishing negativism might be inferred
as features characterizing the internal representational
model of the relationship with the mother in childhood
for those mothers who exhibited hostile attitudes toward
parenting and their own children. As with the parenting
style measure, low self esteem was correlated with
Hostile-Objectification, although it did not contribute
to additional control of the variance. Further,
psychiatric hospitalization during pregnancy or the
child's first year also correlated with Hostile-
Objectification. These relationships did not supplement
the prediction explained by the historical variables:
they may represent exacerbation of psychological
difficulties at times when the representation of the
mother-child relationship is keenly experienced.

Aspects of both negative control and positive
involvement experienced in the relationship with one's
mother in childhood contributed to the prediction of
current parenting attitudes of Overprotection-
Overcontrol. These two contradictory dimensions of the
recollections of parenting one experienced in childhood
perhaps indicate confusion, uncertainty, or ambivalence
emanating from the internal representation of the
relationship with the mother in childhood. This
interpretation is consistent with the findings of Main

and her colleagues (Main, Kaplan, & Cassidy, 1985: Main

113

& Goldwyn, 1984) who found distortion, incoherence, and
disorganization in the childhood attachment
representations of mothers who had insecure attachments
with their own children. Main reported that some of the
mothers who had experienced rejection from their own
mothers also idealized them. This pattern appears
analogous to the present mothers who acknowledged both
positive and negative aspects in their recollections.
These opposing qualities also appear to be incorporated
in the Overprotection-Overcontrol measure,
characterizing the mother's current parental attitudes
toward her own children. Caring aspects toward the
child, reflected through concern for the child's welfare
and proper socialization, are fused with subtly
rejecting attitudes through ignoring of the child's
individuality and autonomy. It should be noted that a
diagnosis of schizophrenic disorder contributed further
to the prediction of Overprotection-Overcontrol,
suggesting that the disorder also may augment the merger
of the contradictory elements in these parenting
attitudes.

Finally, perceived difficulty or uncertainty in the
parenting role was also predicted by perceptions of
negative control from one's own mother in childhood.
Thus, the representation of the relationship with one's

own mother as negativistic and punitive was related to

114

both of the problematic types of parenting attitudes as
well as the perception of stress in carrying out the
maternal role.
Education and Parenting

Beyond the measures of parenting experienced in
childhood, Hostile-Objectification and Overprotection-
Overcontrol also were predicted by the mother's
educational level. Similarly, education also
contributed to the prediction of feelings of difficulty
or uncertainty in the parenting role. Educational
attainment (40% did not finish high school) may provide
an index of the level of social adjustment achieved by
the disordered mothers. In this light, mothers with
severe psychopathology who were more limited in the
adequacy of social adjustment they had ever achieved
were more likely to exhibit detrimental parenting
attitudes and difficulty in the parenting role. Younger
age at first birth was related to higher Overprotection-
Overcontrol, supplying additional evidence for less
adaptive social adjustment being related to
inappropriate parenting attitudes. However, age at
first birth did not contribute to the regression because
age at first birth was strongly related to education.
In addition to education, younger age at first birth was
predicted by younger age of first psychiatric

hospitalization. Dropping out of school and having a

115

child as well as experiencing a psychiatric
hospitalization at a younger age suggest various aspects
of poorer social adjustment which are likely to
contribute to problematic parenting. Social adjustment
may be more important than severity of the disorder: the
parenting attitudes were not significantly correlated
with measures indicative of the severity of disorder
(e.g., total number of psychiatric admissions, number of
hospitalizations and number of days hospitalized in the
last three years).
Physical Punishment

Unlike the other parenting measures, the use of
physical punishment as a means of responding to and
disciplining a child was not related to the mother's
childhood experiences. Those mothers who were single,
black, and young were more likely to prescribe physical
punishment as a means to control their children as were
those mothers who received public assistance.
Collectively, these features characterize a subgroup of
the disordered mothers who culturally would appear most
disadvantaged, even though the whole sample was
generally of low socioeconomic status. These mothers
with fewer resources may have less tolerance in dealing
with their children and may use physical punishment as
an attempt to rapidly eliminate a perceived additional

source of stress and frustration in their lives.

116

Alternatively, as the physical punishment measure was
correlated with Overprotection-Overcontrol, the more
disadvantaged mothers may be overly concerned about
transgressions and improper child behavior and apply
extreme measures to regulate their children's behavior.
The majority of mothers (77%) included their children as
social network members, a finding also reported by Belle
(1982). However, not including one's children as
network members also contributed to the prediction of
the use of physical punishment. This suggests that the
mothers who used physical punishment may view their
children differently than other mothers, perhaps
regarding them as objects to be regulated or potential
sources of stressful demands for instrumental
functioning.
Potential Sources of Stress and Support in Parenting
Sources of stress in the mother's life may tax her
functioning as a parent. The mother's perception of
behavior problems in her child was assessed as an
indication of potential stress for the mother (rather
than as an accurate measure of child adjustment).
However, the perception of child difficulty was not
related to any of the parenting measures. Predictors of
greater perceived child behavioral difficulty included
negative control experienced in the relationship with

the mother's own mother, adequacy of the living

117

environment, and number of network supporters providing
childcare assistance. As previously discussed, the
representation of a punitive, negativistic relationship
with one's own mother appears to correspond with
perceived difficulty in the relationship with one's own
children. There appears to be consistency in the
mother's internal model of the parent-child relationship
applied to different relational contexts. The model
would be one in which the ”mother" views her "child" as
a source of aversion. The participant in the role of
the child in relationship to her own mother infers that
the ”child" must be difficult or bad in order for the
"mother" to respond with negative control. The same
relationship is conferred on the current mother-child
relational context in which the "mother" views the
”child” as difficult.

The mother's perception of her living environment
as less adequate or more stressful predicted the report
of more behavioral disturbance in the child. From a
perceptual perspective, if a mother feels that the
conditions and available resources in a number of areas
of her home life are inadequate, restrictive, or
unsatisfying, she is more likely to feel helpless or
stressed and may generalize these perceptions of
environmental strain to include the behavior of her

child. Alternatively, the mother's report may reflect a

118

more realistic appraisal of environmental deprivation
and inadequacy which contribute to less Optimal
development in the child. Features such as poor
financial resources, awkward or troubled living
situation, dangerous neighborhood, etc. indicate
environmental stresses which would impinge upon optimal
development.

Contrary to expectations, having more network
members who provide support in the form of childcare
assistance or advice predicted more child behavioral
difficulty. It had been expected that if a mother had
more individuals on whom to rely for assistance in
childcare or advice related to childcare issues, then
she would feel more supported in her parental role,
promoting adjustment in the child. However, the
negative relationship may imply that a mother is more
likely to seek assistance from others if her child is
more difficult to manage. Alternatively, a mother, who
has many individuals caring for her child, may expose
her child to a confusing, inconsistent array of
caregiving situations, promoting behavioral difficulty
in the child. Further, multiple caregivers in the
network may signify a strain in the attachment
relationship between the mother and her child as a
result of a tendency for the mother to abdicate

childcare responsibilities or less commitment to the

119

caregiver role.

The descriptive characteristics of the sample of
disordered mothers as a group indicate numerous factors
which are likely to promote stress and risk in parenting
beyond the impact of psychopathology alone. Although
the mother's psychopathology is likely to contribute to
some of these factors, these factors are likely to have
an additional impact on her parenting ability and the
development of her children. Low income, limited
education, and considerable long-term unemployment
indicated few resources were available to support the
parenting context. Many of the mothers first gave birth
as teenagers, and the majority had given birth to at
least one child while unmarried, indicating poor
preparation for the demands of parenting. Only 21% of
the mothers were married, indicating that very few of
the mothers were parenting with the support of a spouse.
Of those mothers who had ever married, 69% had
experienced at least one marital disruption. Half of
the mothers (51%) did not regard their neighborhoods as
safe places to live. Collectively, these features
depict a parenting context that is likely to be
extremely taxed, irrespective of the presence of
psychopathology.

An extended kin support network appeared to be one

resource enhancing the lives of most of the mothers.

120

The social support networks of the mothers tended to be
small and dominated by kin. In comparison with mothers
in a community sample of middle class mothers assessed
with a similar social support questionnaire (G. A.
Bogat, personal communication, June, 1987), the number
of social supporters in the disordered mothers' networks
was smaller than that of the community mothers (M§11.5
vs. M§l9.0), and their networks were dominated by
relatives (66% vs. 32%). These differences resulted
from the disordered mothers having fewer nonkin
supporters than the community mothers, while both groups
had a similar number of relatives in their networks.

The disordered mothers may have had fewer nonkin in
their networks as a consequence of social withdrawal
associated with their psychopathology, whereas relatives
would be more likely to be retained in the network due
to kinship ties rather than active effort from the
mothers. Alternatively, the mothers may be limited in
the nonkin ties they are able to establish because, as a
result of their disorders, they are less capable of
engaging in mutual reciprocation of support, a pattern
typifying support networks in poverty samples (Belle,
1984). If it is assumed that disordered mothers are
less able to engage in reciprocal provision of support,
their kin supporters may feel burdened by having to

provide support but not receiving support in return.

121

The mothers' psychopathology and recurrent psychiatric
hospitalizations may also stress the lives of their kin
supporters through the supporters' distress and concern
over the mothers' condition and by the added demands of
caring for the mothers' children during hospitalization
(Cohler, 1984). In turn, this may diminish the quality
of the support the mothers are able to obtain from their
networks, hampering their community adjustment and
parenting role.

Research Limitations and Future Directions

The findings of the present investigation need to
be replicated with a community sample of mothers.
Interviewing these women in the hospital while they were
separated from their children and removed from the
responsibilities of parenting may have influenced the
results. Reassessment of hospitalized mothers after
they have returned to their families and comparison with
the current results would aid in evaluating the
reliability of the findings.

In attempting to assess a broad range of possible
influences on parenting, many measures were used.
Although efforts were made to shorten the interview, the
length of time needed for completion was quite long, and
it was difficult for many of the mothers to maintain
consistent concentration. With the goals of

comprehensiveness and brevity, some measures were

122

selected because of their short length, although they
did not exhibit Optimal psychometric properties or did
not have extensive or apprOpriate norms, thus, limiting
interpretation. Some of the negative results found in
the investigation may have resulted from methods that
did not adequately assess the construct of interest.
Further, as all information was obtained from the
participant or hospital records, the accuracy of some
material may be questioned. This problem could be
alleviated to some degree by incorporating interviews
with other family members. Also, as the interview
measures with the mother were largely of self-report
format, it is difficult to determine the extent to which
the responses reflected actual circumstances. Future
use of observational measures, such as rating scale
schedules to evaluate the home environment or
observations of the mother in interaction with her
children, would aid in more accurately assessing
components of the conceptual model. Independent
evaluation of the adjustment of the mother's children is
a critical component that was not possible in this
investigation.

Although the mother's accounts of relationship
qualities in her childhood assumed significant
prominence in this investigation, the perennial problems

with retrospective reports must be considered: the

123

extent to which the mothers reports were accurate
indications of childhood experiences is not known.
Nevertheless, her responses may reflect her current
representation of those experiences which are more
likely to influence her parenting at present. Repeating
the measurement of these constructs when the mother is
not hospitalized would help to determine the reliability
of her report. Also, to restrict the length of the
interview, the mother was only asked to depict her
relationship with her mother: additional investigation
of the qualitative aspects of the participant's
relationship with her father would provide a broader
perspective on her relational experiences in her family
of origin.

Finally, the small sample size necessitates
replication with a larger sample for greater confidence
in the results reported. Although the sample reflected
the demographic characteristics of mothers who were
hospitalized in state psychiatric facilities to a
reasonable degree, generalization of the findings to
groups of more diverse socioeconomic levels and of more
representative racial and marital composition may be
limited. The general social deprivation of this sample
makes it difficult to determine the extent to which
parenting patterns are a result of psychopathology or

cultural deprivation. However, this may be a strength

124

of this study: it may more accurately depict the
desperate situation that many mentally ill women face in
parenting their children than do studies that recruit
participants who are middle class and white with intact
marriages (cf., Baldwin, Cole, & Baldwin, 1982).
Treatment Implications

Although as a group the disordered mothers may
exhibit less effective parenting than a nondisturbed
comparison group, there was considerable variation
within the disordered group: some mothers were likely to
be more effective parents than others. All of the
mothers might benefit from parent education to foster
competence in the parenting of their children, but some
of the mothers were in more serious need than others.
The effectiveness of teaching or enhancing parenting
skills is likely to be mediated by the mother's ability
to make use of the training. Skills alone may not be
enough: factors contributing to the individual
differences in parenting among the mothers also need to
be addressed. The proximal path analyses suggest
several current life factors that might be targeted for
intervention to improve the mother's parental
functioning. Ecological interventions to improve
emotional support, supportive psychotherapy to improve
self esteem, and intervention to prevent relapse might

enhance the mother's responsiveness to her children.

125

However, the results of the path analyses of distal
models indicated that the mother's childhood relational
or attachment experiences influenced her self esteem and
emotional support. Psychotherapeutic efforts to alter
the major postulates of the mother's internal
representational model of attachment relationships may
contribute to direct improvement of the mother's
responsivity to her children and provide indirect
influences on parenting through improving self esteem
and the mother's ability to engage in emotionally
supportive relationships. Although this apparently is
an important aspect to improve, the task is likely to be
difficult. Nevertheless, it is important to note that
Main and Goldwyn (1984) reported that some mothers who
had experienced negative attachment relationships in
childhood had been able restructure their understandings
of these relationships in later years and exhibited
secure attachments with their children. Thus,
alteration of the internal representational models which
develop in childhood is possible and advantageous for
the mother's effective parental functioning.

In addition to specific interventions to support
the mother in her maternal role, intervention efforts
with the mother's extended family, her major source of
social support, also would be beneficial. These network

members are likely to be stressed by the continuing

126

difficulties of the mother, and assisting them in their
relationships with her might improve the quality of the
social support they are able to provide. Further, as
multiple rehospitalization was shown to be a strongly
negative predictor of the mother's parenting,
intervention with the extended family members might also
help to reduce their potential negative expressed
emotion toward her, a variable which has been shown to
contribute to relapse (Leff & Vaughn, 1985). Assisting
the mother in developing nonkin supportive ties may also
be beneficial: participation in group sessions with
other mothers might be a means to establish mutually
supportive nonkin relationships.’

During hospitalization, the issues relating to
parenthood for the current participants did not appear
to a goal of treatment: in only 19% of the participants'
treatment plans was some mention made of their roles as
mothers. Attempts to aid these women in their maternal
roles could begin while they are hospitalized. Group
meetings of mothers could be developed in order that the
hospitalized mothers could share their experiences of
parenting, concerns over separation, etc. This could
help to establish improved parenting as a goal for the
mothers and form a bridge to aftercare treatment in
which support with parenting issues and role functioning

as a mother could be continued. Specialized staff

127

members in the hospital also could be designated to
attend specifically to the special needs of women who
are mothers and to promote continuity in aftercare
planning for the parenting role of these women.
Finally, given that many of the mothers had been
hospitalized before bearing a child and that many of the
mothers seemed poorly prepared for the responsibilities
of parenting (teenage birth, not married), preventive
efforts in the form of discussions about birth control,
choice in sexual relationships, the demands of
parenting, etc. might foster greater control among
potential mothers in their decisions to have children.
Finally, interventions to assist these women in
their roles as mothers would benefit not only the
mothers' psychological health and development but also
that of their children. It should be noted that some of
these mothers may lose custody of their children in the
future as a result of an inability to continue parenting
or parenting in a destructive or neglecting manner.
Interventions to improve parenting could prevent adverse
effects to both the mothers and their children before
adverse psychological consequences occur and the
situation deteriorates to such an extent that more

stringent measures are required.

APPENDICES

APPENDIX A

CURRENT LIVING SITUATION QUESTIONNAIRE

128

Current Living Situation Questionnaire
”I'd like to begin by asking you some questions about
your life before you came to the hospital."
First, what is your marital status? Are you:
Single - Married - Separated - Divorced - Widowed
(If ever-married), How old were you when you first

married?

(If single,) are you involved in an intimate

relationship? (Do you have a boyfriend or close male
friend?)

If yes, does this person live with you? yes no
(If married,) how long have you been married?

How old is your husband?

Have you ever been married before this current

marriage?
If yes, describe (number of marriages, how long):

(If separated,) how long have you been separated?
How long had you been married?

Had you ever been married before this marriage?
If yes, describe (number of marriages, how long):

(If divorced,) how long have you been divorced?
How long had you been married?

Had you been married before your last marriage?
If yes, describe (number of marriages, how long):
(If widowed,) how long have you been widowed?

How long had you been married?

Had you been married before your last marriage?
If yes, describe (number of marriages, how long):

129

Education: How many years of school have you completed?

Education of spouse (if applicable):

Employment: What was the last job you had?
Full or parttime? How long?

How long ago was it that you last worked?

Employment of spouse (if applicable):

Is your husband employed?

If yes, what type of work does your husband do?
Full or parttime? How long?

Now I'd like to ask you some questions about your
income.
Do you receive G.A. (welfare)?
Do you receive aid for dependent children (ADC)?
Do you receive food stamps?
Do you receive SSI/SSDI?

How well does the amount of money your family has take
care of what you and your children need?

Very well - fairly well - varies - poorly - very poorly

130

Children: Now I'd like you to tell me some things about
your children.
For each child, ask the following:

Child # l 2 3 4 5 6
Name
Sex

Age &
Birthdate

Lived w/ you
before hosp?

How long had
you lived w/
before hosp?

Who cares for
while in hosp?

Now I'd like to ask you some questions about your
pregnancy and the first year with each of your children.
(Transfer the names of the children and complete the
following chart.)

131

 

 

 

 

 

 

 

Topics-v eons uHJ

as so»

couuaano use» no as. no
swan mo hash a-uuu “mm
acumen eludpoun Ounce
ranches mucus no scan»
eanaHauan-On oauauaco
ease ace o>sc so» can

 

A.¢p«uonoc 3a.» may

0: no»
hcohuHaso use» «o and
5:. g a... so»
can: escapes: can»-
easo»nn Hesse no need»
esuHHsuHA-o: Ouuasaco
sang and use: so» can

 

4

 

A.opduosoc em.» adv
o: no» ecouquco use»
co 2. 5:. .3. .33

2538.. £83 33828

329.3 3.: 3338:.—
usou sends has no and:
sucks one: so» can: OIOH
snows use use: so» can

A

can. you usoaosv
2.5... 3323 .5

 

.OHDdOHHA—Qd NH 8... Cd
can: usages: scanned

$8251.03. .2 .2
7328 .2 3 :85: 2.88

oh; 33:33 .0: u:
o: no» whores» cede
0:» ass: couuadzu

 

coccdans no accuses
hocdcnoua use» as

a...» no 2. Ba

 

one: 3 «no

132

Have you ever had a miscarriage? yes no
If yes, describe time(s) and ask: Did you have a
psychiatric hospitalization around that time?

Aside from psychiatric hospitalizations, have there been
other times when you were separated from your children
for a week or more (i.e., if custody was temporarily
lost, if someone else was caring for children for a
period of time, if children in foster care, etc.)? yes
no

If yes, describe:

Have you ever lost any of your children (i.e., as a
result of adoption, child dying, etc.) yes no
If yes, describe:

Overall, when you are not in the hospital, how do you
feel about being a mother? Are you:
very satisfied
satisfied
in between
dissatisfied
very dissatisfied

How able are you to take care of your children when you
are not in the hospital? Is it:
very easy
easy
in between
difficult
very difficult

133

Now I'm going to ask you some questions about where you
were living before you came to the hospital.

What kind of place were you living in?
apartment - duplex - trailer - house - other(specify):

Do you rent( ) or own( ) or
Is it a friend's or relative's place that you share
() If checked, whose?

Who are all the people that lived there? (List persons
and their relationship to mother)

How many bedrooms were there where you were living?

Was there enough space in your home?
very roomy roomy so-so cramped very cramped

How long had you lived there?
If less than 1 year,
how many times have you moved in the past year?
(Describe):

Did your home have a telephone?
Do you own a car? If no, do you have access to one?

How many blocks was it to the bus from where you were
living?
How often did you ride the bus?
very often-fairly often-sometimes-not very often-never

How easy or difficult was it for you to get places you
needed to go?
very easy somewhat easy depends

difficult very difficult

How satisfied were you with your living situation?
Very - satisfied - neutral - dissatisfied - very
satisfied dissatisfied

How safe was the neighborhood where you were living?
very safe - safe - so-so - unsafe - very unsafe

What sort of place was your neighborhood for children to
grow up in?
very good - good - in between - bad - very bad

APPENDIX B

EARLY LIFE QUESTIONNAIRE

134

Early Life Questionnaire

Now I'd like to ask you some questions about what it was
like for you when you were growing up.

Did you live with both of your parents the whole time
you were growing up (until 18)? Yes No

If you did not live with both of your parents the whole
time, why not? (check reason and age when it
happened)
Age
)death of mother
)death of father
)divorce or separation of parents
)economic difficulties
)schooling or institutions
)always lived with someone else
)military
)prison
)other (specify):

AAAAAAAAA

If you did not live with both of your parents the whole
time, please describe who you lived with, and any
changes in living arrangements that occurred as you were
growing up? (for each change): How old were you when
that change happened?

Now I'd like to know about any times you were separated
from your mother when you were growing up. Were you
ever separated from your mother for a week or more as
you were growing up? yes no

If yes, please describe each time, how old you were, and
for how long.

Were you ever separated from your father for a week or
more as you were growing up? yes no

If yes, please describe each time, how old you were, and
for how long.

135

How many times did you make a move before you were 18?
Please describe. (If not covered on previous page.)

How many brothers and sisters did you have?
Older or younger?
#brothers #older #younger
#sisters #older #younger

Please rate the quality of your relationship with (how
well you got along with) the people listed below
while you were growing up:

very - reasonably - so-so - somewhat - very

good good poor poor
2 3 4 5
Mother 1 2 3 4 5 (n/a)
Father 1 2 3 4 5 (n/a)
Siblings 1 2 3 4 5 (n/a)
Grandparents l 2 3 4 5 (n/a)

Other relatives living
with you(specify):
l 2 3 4 5 (n/a)

1 2 3 4 5 (n/a)

As a child was there someone to whom you felt
particularly close?

yes no

If yes, who? (list up to 3 names and their

relationship in order of preference)

1.

2.

3.

How well did your parents get along while you were
growing up?
very - reasonably - so-so - somewhat - very
well well poorly poorly
l 2 3 4 5

136

(If parents were remarried at some point during
patient's childhood, ask the patient to also rate the
relationship of that marital pair if she lived with
them. Also circle which of the persons she is
indicating.)
How well did your mother/father and your
stepfather/stepmother get along?

very - reasonably - so-so - somewhat - very

well well poorly poorly

l 2 3 4 5

Now I'd like to know if any members of your family had
any psychological or emotional problems while you were
growing up. Did your mother have any psychiatric
problems? Was she ever hospitalized? If so, how many
times? (Repeat these questions for other members).

Psychiatric Ever hospitalized

Problems (Psychiatric)
Mother yes no dk n/a yes no dk n/a
Father yes no dk n/a yes no dk n/a
Siblings yes no dk n/a yes no dk n/a
Grandparents yes no dk n/a yes no dk n/a
Other realtives
Who: yes no dk n/a yes no dk n/a

21

Who: yes no dk n/a yes no n/a

APPENDIX C

CHILD REPORTS OF PARENTAL BEHAVIOR INVENTORY

137

Child Report of Parental Behavior Inventory

Now I'd like to ask you some questions specifically
about what you feel your mother was like when you were a
child. I'll read you a statement and you tell me how
true it was of your mother using the following scale:

TRUE OF MY MOTHER 1
SORT OF TRUE OF MY MOTHER 2
NOT TRUE OF MY MOTHER 3

1. Made me feel better after talking over my worries
with her

2. Often gave up something to get something for me
3. Believed in showing her love for me

4. Kept reminding me about things I was not allowed to
do

5. Enjoyed it when I brought friends to my home
6. Gave me as much freedom as I wanted
7. Understood my problems and my worries

8. Made me feel like the most important person in her
life

9. Worried about me when I was away

10. Believed that all my bad behavior should be punished
in some way

11. Always listened to my ideas and opinions

12. Asked me to tell her everything that happened when I
was away from home

13. Thought I was not grateful when I didn't obey
l4. Didn't forget very quickly the things I did wrong

15. Didn't pay much attention to my misbehavior

16.

17.

18.

19.
20.

21.

22.
23.
24.
25.

26.

27.

28.

29.
30.

31.

32.
33.
34.

35.

36.

37.

138

Allowed me to tell her if I thought my ideas were
better than hers

Excused my bad conduct

Said some day that I would be punished for my bad
behavior

Let me go any place I pleased without asking
Comforted me when I was afraid

Enjoyed staying at home with me more than going out
with her friends

Became very involved in my life

Often praised me

Told me how much she had suffered for me
Wanted to control whatever I did

Let me help decide how to do things we were working
on

Let me stay up late if I kept asking

Said that someday I'd be sorry that I wasn't a
better child

Seemed proud of the things I did
Made her whole life center around her children

Told me where to find out about things I wanted to
know

Asked other peOple what I did away from home
Was always trying to change me
Seldom insisted that I do anything

Asked me what I thought about how we should do
things

Thought that any misbehavior was very serious and
would have future consequences

Didn't show that she loved me

38.
39.
40.

41.

42.
43.

44.

45.
46.

47.

48.

139

Allowed me to spend my money in any way I liked
Spent almost all of her free time with her children
Said I made her happy

Almost always wanted to know who phoned me or wrote
me and what they said

Told me of all the things she had done for me
Didn't let me decide things for myself

Let me get away without doing work I had been given
to do

Gave me the choice of what to do whenever possible
Could be talked into things easily

WOuld talk to me again and again about anything bad
I did

Let me do anything I liked to do

Indicate whom the patient described:

1. Natural mother

2. Mother substitute (specify):

APPENDIX D

SYMPTOM CHECKLIST - 10

140

Client-Self Evaluation (SOL-10)

The next thing I'd like to do is ask you about some

problems that people sometimes have.

I'm going to ask

you a question and I'd like you to tell me HOW MUCH
DISCOMFORT THAT PROBLEM HAS CAUSED YOU DURING THE PAST

WEEK INCLUDING TODAY.

the question applies to you.

Use this scale to let me know how

ExtremelyHQuite a bit-Moderately-A little bit-Not at all

1 2

1. How much were

2. How much were
in things?

3. How much were

open spaces or on

4. How much were
of your body?

5. How much were

6. How much were
your arms and

7. How much were

3

you distressed

you distressed

you distressed

you distressed

you distressed

you distressed
legs?

you distressed

out of the house alone?

8. How much were
keyed up?

9. How much were

worthlessness?

10. How much were
when you were

you distressed

you distressed

you distressed
with people?

by
by

by

the streets?

by

by

by

by

by

by

by

4 5

feeling lonely?

feeling no interest

feeling afraid in

feeling weak in part

feeling blue?

heavy feelings in

feeling afraid to go

feeling tense or

feelings of

feeling lonely even

APPENDIX E

SELF ESTEEM SCALE

141

Rosenberg Self-Esteem Scale

 

Now I'd like to ask you some more questions about how
you feel about yourself. I'll read a statement to you
and I'd like you to tell me how much you agree or
disagree with the statement using the following scale:

1 2 3 4
Strongly Disagree Agree Strongly
Disagree Agree

1. I feel I'm a person of worth, at least on an equal
plane with others.

2. I feel that I have a number of good qualities.

3. All in all, I am inclined to feel I am a failure.

4. I am able to do things as well as most other peOple.
5. I feel I do not have much to be proud of

6. I take a positive attitude toward myself

7. On the whole, I am satisfied with myself.

8. I wish I could have more respect for myself

9. I certainly feel useless at times.

10. At times I think I am no good at all.

APPENDIX F

SOCIAL SUPPORT QUESTIONNAIRE

142

Social Support Questionnaire

 

Now I'm going to be asking you some questions about
peOple who are part of your life and provide you with
help or support. There are nine questions. For each
question, think of the people in your life before you
came to the hospital who provided you with the kind of
support asked for. Tell me the first names of those
peOple and I'll write them down. If you have two peOple
with the same first name, then give me the first initial
of their last name so I can tell them apart. You can
name the same person as often as you want. Finally, if
there is no one who helps you in a certain area, just
tell me so.

1. Who do you spend time with? (such as, get together
for a cup of coffee or dinner, enjoy talking with,
etc.?)

1. 6.
20 70
3. 8.
4. 9.
5. 10.

2. Who do you go out with for social or recreational
activities? (such as, going to movies, going out to eat,
going to clubs or church groups, or other things like
that?)

UlubUNI-i
o 0 0
0‘9me
0 o o o

0
g...

3. Who gives you information and advice when you have
problems with your children? (such as when your child is
sick, when your child misbehaves, when you don't know
how to handle your child, etc.)

U'lebWNl-H
o 0
owmflm
o o o

o
H

143

4. Who do you go to for information or advice for such
matters as handling money problems, finding a new job,
making household repairs, solving personal problems?

1. 6.
2. 7.
3. 8.
4. 9.
5. 10.

5. Who helps you with taking care of your children?
(such as babysitting, helping out if your child is sick,
helping if an emergency occurred with your child?)

1. 6.
2. 7.
3. 8.
4. 9.
5. 10.

6. Who helps you with matters of daily living? (such as
giving you a ride somewhere, letting you borrow a little
money if you needed it, helping you get something
repaired, helping you with housecleaning, laundry,
cooking, etc.?)

1. 6.
2. 7.
3. 8.
4. 9.
5. 10.

7. Who can you confide in when you have a personal
problem or when you feel upset?

I. 6.
2. 7.
3. 8.
4. 9.
5. 10.

8. Who cares about you?

U'IQUJNH
o

ommflm

o o o o o

o
g...

144

9. Who makes your life difficult? (such as someone who
expects too much from you, makes too many demands on
you, someone who you wish would leave you alone or
someone who you would like to avoid, etc.?)

0 6.
7.
8.
9.

10.

U'IwaH

Next, I'm going to ask you some questions about each of
the persons you mentioned on these questions. I'm going
to write all of their names down. For each person, I
want to know their sex, their relationship to you, how
often you have contact with the person when you are not
in the hospital, and how satisfied you are with your
present relationship with that person.

(After entering in all of the names, proceed through
each individual supporter and ask the questions about
each one. For the relationship question, have the
patient indicate all of the relationships that apply.)

145

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m e m N a u N u a u o < N A z c a a a _ a a N u a u a < N a .4.
n e m N n u N u a u a c N a z a a a a a a u a a a u a < N H .m.
m e m N _ o N a o u a c N _ z c 4 x a _ n u m u o u a c N _ .N.
n c n N a u a u a u a < N a z c a x a _ a u N u a u a < N a .«a
h e N N _ u N u a u a c N _ z n a a a a a u N u a u a c N 4 .o.
n v m N _ o N u a u a < N _ z z a s a _ = u N u a u a c N ~ .3
nmva_ Umuoucc N. z:...a..—=U..aoum< NH .o
h v m N a u N u o u a < N _ z n a x a — a u N u a u a < N a .N
r v r N N o L m a u a < N u z z a a a _ z u N u a u.m < N H .m
n v m N _ U m u a u a < N _ z t a a a _ a U N m a u a < N _ .m
m e m N a o a u c u a c N a z c a x a _ z u a u a u a < N ~ .4
m e m N a o a a a u a < N _ z c s x a _ a u N u a a a < N a .n
a v m N a U N u a u a < N . z c a a a u = U N a a u a ¢ N a .N
m e m N a u a m a u a c N L z z a x a _ a u N u a u a c N a .q

aosuo.z

accenuuou0ua.:

cuoHocoa.a

accused uozno.s

Loxuoanou.u

censuses.—

cco..m.=

hand no coax c 00:0.0 Nassau Lognonu

enoxxmusau som.m ocuenaucaeo.m

ccqaaascu Neu>.m anco- a 00:0.» aco~«50.u

sesamauam.e nucOnxaueuu aum.e hwmum.o

cuuauos aim gooseoUcod a Name 0 .
cu.~m.ueam.o.N gooaxuc-«u :um.a 02 .N cause“ om.a u—a- N.N
uuqamanamudu Neu>.~ Nee Nousu.< no» .A unsoau\0uc: 4 one: a
Neucang and: nuanceso «denuded: ca soc cut: «:30 so cueesngu adgacosuaqux xom u-az

Isaac scanned Lao» nus: coax-«g zuaa Dungeon
:0» use unquauuaa so: o>ac so» On guano so:

o>an census moo:

APPENDIX G

MOTHER-CHILD RELATIONSHIP EVALUATION

146

Mother-Child Relationship Evaluation (MCRE)

 

The next set of questions I am going to ask you are to
better understand you and your child and your
relationship with your child or children. I am going to
read a statement to you and I would like you to express
your feelings or opinions about the statement as it
pertains to you as a mother. Use the following scale to
express your answer:

Strongly Strongly
Disagree Disagree Undecided Agree Agree
1 2 3 4 5

There are no "right" or ”wrong" answers, only your
opinions and feelings. Let your personal experiences
decide your answers.

1. If possible, a mother should give her child all
those things the mother never had.

2. Children are like small animals and can be trained
the same as puppies.

3. Children cannot choose the proper foods for
themselves.

4. It is good for a child to be separated from his/her
mother from time to time.

5. ”Having fun" usually is a waste of time for a child.
6. A mother should defend her child from criticism.

7. A child is not at fault when he/she does something
wrong.

8. When a mother disapproves of an activity of her
child, she should overemphasize its danger.

9. My child cannot get along without me.

10. My child does not get along with other children as
well as he/she should.

11. A mother should be resigned to the fate of her
child.

12. A mother should see that her child's homework is
done correctly.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.
29.
30.

31.

32.

147

To raise a child suitably, the mother should know
fairly well what she would like her child to be.

A mother should “show off" her child at every
opportunity.

It takes much energy to discipline a child properly.

A mother should never leave her child by
him/herself.

With the right training, a child can be made to do
almost anything.

It is good for a mother to cut her child's hair if
he/she dislikes going to the barber.

I often threaten to punish my child but never do it.

When a child disobeys in school, the teacher should
punish him/her.

My child annoys me.

It is the mother's responsibility to see that her
child is never unhappy.

A child is an adult in small form.

A mother cannot spend too much time reading to her
child.

A child needs more than two medical examinations
each year.

Children cannot be trusted to do things by
themselves.

Breast feeding should be stopped by the mothers as
soon as possible.

Children should always be kept calm.
A child should not have a fixed allowance.
I often play practical jokes on my child.

The mother should lie down with her child if he/she
cannot sleep.

Often children act sick when they are not sick.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.
45.

46.

47.

48.

148
Children can never bathe themselves as they should.

A child should not be scolded for grabbing things
from an adult.

When a mother has problems with her child with which
she cannnot deal, she should seek the proper help.

When a child cries, he/she should have the mother's
attention at once.

Somehow, I cannot refuse any request my child makes.
Children have rights of their own.

A mother should always see that her child's demands
are met.

A child should not get angry at his/her mother.

Young children, like toys, are for their parents'
amusement.

Childbearing is a responsibility of marriage.

There are certain right ways of raising a child, no
matter how the parents feel.

Children should be seen and not heard.
A mother should control her child's emotions.

Since thumbsucking is an unhealthy habit, it should
be stOpped by all means. '

It is not too helpful for a mother to talk over her
plans with her child.

A child should please his/her parents.

APPENDIX H

PARENT ATTITUDE SURVEY SCALES

149

Parent Attitude Survey (Difficulty Scale)

The next group of questions also are about how you feel
about being a parent. Use the following scale to
express your feeling about each statement:

12.

13.

14.

l 2 3 4
Strongly Disagree Agree Strongly
Disagree Agree

I feel I am faced with more problems than most
parents.

Few parents have to face the problems I find with my
children.

It's hard to know what to do when a child is afraid
of something that won't hurt him.

Children don't realize that it mainly takes
suffering to be a good parent.

Parents sacrifice most of their fun for their
children.

Raising children isn't as hard as most parents let
on.

It's hard to know when to make a rule and stick by
it.

Raising children is a nerve-wracking job.
It's hard to know what healthy sex ideas are.
A parent has to suffer much and say little.

It's hard to know whether to be playful rather than
dignified with children.

It's a rare parent who can be even-tempered with the
children all day.

Fewer people are doing a good job of child-rearing
now than 30 years ago.

Taking care of a small baby is something that no
woman should be expected to do all by herself.

APPENDIX I

SENSITIVITY TO CHILDREN QUESTIONNAIRE

150
S T C Questionnaire

The next thing I'd like you to do is a little different.
I am going to describe a number of situations to you.
You are to pretend that you are the mother of the child
described. All of the children in the situations are to
be considered four years old. Your task is to describe
exactly how you would respond to the child in each of
the situations. You should say what your exact words
and actions would be as if it were a script.

1. You have been feeling very depressed and wish there
was someone you could confide in about your troubles.
Your daughter, Tina, comes running into the house and
says, in a quivering, sobbing voice, ”Some kids made fun
of me. They said I was a baby and that I couldn't play
with them."

Examples:

Authoritative (38.3%)

I'd say, “I don't understand why peOple do stuff
like that. Why they said that. Maybe they won't be
like that next time. You don't like to be called a
baby. That makes you feel bad.”

Indulgent (23.4%)
Say, ”You're Mama's big girl."

Authoritarian (10.6%)

I would ask her, ”Were you acting like a baby? You
have to learn other kids aren't as patient as your
mother."

Neglecting (19.1%)
I'd go in the house and sit around and do nothing.

Psychotic Intrusion (8.5%)

"Don't cry, don't cry cause you'll have friends one
day. Look I'm not your keeper. It's time for you to
get out and fend for yourself. And get to work and
learn your alphabet.” Get some cards, and we'd make up
our work, and they'd sit and watch me. Never be without
paper and pencil.

151

2. You've been under a lot of pressure lately and as you
are rushing to get dinner ready, there is a sudden loud
yelling in the living room of your home. You go into
the room and you see your daughter, Martha, pushing her
two year old sister, Sarah. Sarah is crying.

Examples:

Authoritative (40.4%)

I would stop Martha from pushing Sarah and ask,
”What's going on between you two?" Try to see if the
problem can be solved. I would see what I could do to
straighten out the problem.

Indulgent (4.3%)

”Sarah, why are you crying? May I help you
please?” Ask Marth why she was crying. I'd approach
Sarah because she was the one who was crying.

Authoritarian (29.8%)

I would spank Martha and send her to her room and
make her stay there the rest of the day. I'd say, "Why
did you do it?" Her reason, whatever, would be
insufficient, and she'd be punished. Any argument
between the two should be brought to me.

Neglecting (19.1%)
I'd let her wOOp Sarah.

Psychotic Intrusion (6.4%)

”Get up and get some milk. Get up and change these
clothes. Get up and get hygiene." Then I tell you, 'Go
to church, not by yourself. Bring your children, too.”

152

3. You are eager to go out for the evening which you
haven't been able to do for a long time. As you and
your spouse are preparing to leave, your son, Sam,
exclaims, "Please don't 90. Please stay home tonight.”
He doesn't appear sick, and the babysitter is a person
he has previously enjoyed being with.

Examples:

Authoritative (31.9%)

"You must really want me to stay. But now I've
made plans, and I've got to go. Maybe later we can do
something.”

Indulgent (23.4%)
I'd stay home. I couldn't go out.

Authoritarian (14.9%)

"I'm going, and you're not going to stop me because
I need some different scenery. I can't be mother,
doctor, nurse, etc. all at the same time."

Neglecting (25.5%)

I'd leave the house. They cried for two minutes
after: then they'd forget about it. At first, it
bothered me, but then I'd realize that they always use
the same tactics. I would just leave.

Psychotic Intrusion (4.3%)
"Do we have to hide our feelings to become man and
wife?”

153

4. You have just finished doing several loads of
laundry. Your son, Peter, comes to you with a big smile
on his face to show you a finger painting he made. The
whole page is covered and brightly colored. His hands,
face, and shirt are also covered with finger paint.

Examples:

Authoritative (21.3%)

I'd admire the picture and tell him to take a bath.
Then I'd put the picture on the wall and tell him next
time to wear an old shirt.

Indulgent (23.4%)
"The picture you drew is beautiful."

Authoritarian (25.5%)

I'd tell him, ”There's no way you can be doing
that. You have to be supervised by someone. Otherwise,
you might destroy things around the house.

Neglecting (25.5%)
Just let him go to bed with it one. I worked today
and will get it off tomorrow.

Psychotic Intrusion (4.3%)

Put the baby in the bathtub. While I sit there
washing up, he can sit and see hisself, and his
curiosity begin the thing. I'd say he'd be trying to
reach for it. I'd say, ”Uh, uh. Get out of the tub and
go in your room and get on your underclothes.”

154

5. You walk into your bedroom and find you son, Burt,
putting your wallet down with a $10.00 bill in his hand.
It is clear from his actions (looking shocked at your
arrival, putting his hand with the money behind his
back) that you have caught him stealing.

Examples:

Authoritative (27.7%)

"If you want money, just ask for it. You shouldn't
be going in other people's wallet cause that's called
stealing. Give me the money back. If you want money
just ask for it."

Indulgent (4.3%)

Ask him, "Why are you taking that out of my purse?
You know I give you basically whatever you need, and
you're probably taking that to give to your friends.
Friends who are less fortunate.”

Authoritarian (51.1%)

Ask, "What are you doing with that $10 bill? What
are you doing in my purse? Stealing is a sin. I don't
believe in thieves. I don't want a thief in my house."

Neglecting (12.8%)
Smack him in the mouth.

Psychotic Intrusion (4.3%)

”You need it, and I need it. I have lived here,
and I've come to pay you a little receipt for your rent,
but thank the Lord that we were not able to hurt anyone
who was staying outside with a gun."

 

APPENDIX J

CHILD BEHAVIOR CHECKLIST

CHILD BEHAVIOR(CHECK IST FOR AGES 4-16

155

 

on I Summon”

 

 

CHLO‘S
wane

PWNI‘SMO'WOM(mum-Umr mm In"
mdwmmmmcm:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

-0 - .. .2 mm
D 3‘" women
rooms one CHILD‘S UM?! nnwm
Ho. 0" n Ila. 00y Yr. MMMOOUIIY
C] m
m 0 new
some. C] omen-cm
I. Mflummmmm Whoa-«Melan- emu-mammoth-
bmpmnwwm “melodies-”Ian mmMflmmmu-a
wmmmm “Hm-munch? “no?
mamas. I...
(:1 Menu M n... W nun can u.- Alon
““" m... Ila- “ W m
C] D C] D U D C] D
a C] C] C] Cl C] Cl C] U
a C] D C] D D Cl Cl C]
I. “Mammal-much“... Whom-(Mdho Whom-toluene!“
mum-“mm “mantle-amen. ”mm-olmhdm“
Forwumdoutmm mum-panacea? mono?
cram. singing. dc. (Do not Include TN.) w I”
Den
0 None lac: M AW 1h- : lob. m A.”
. D D D C] C] D C] D
u C] C] D C] C] C] D D
'- Cl C] D C] Cl C] C] D
II. mum-ems... Whoa-1M0!“
mammefl‘mu “nabs-mandati-
D m code?
M
4 C] U D D
a. U U D U
a D C] C] D
N. mummummm emu-“mum
Mfaummmnwum “mandamus!“
Magnetism. manna-I?
U m m can Aim
Ins- Aw m Avon.
e C] C] D 0
°- [3 D D C]
5 D C] E] C]
omen-.muw-rmctmmn“ note “slal-

156

 

v. e. wmmmemmmmmv D None C]! D 20:: D lume’

2. Amunmeeyflmeeweeteeeeyowemueownoedmmem7 (:1 “semen: D :07? D Jennon

 

VI. Wummummmmmmomm
Worse monsoon» lens:

e. Ge: um was new when 1. Meters? C] C] C)
e 6e: m were nine! W? C] C] (:1

some one new pushes: C] C] C]
4. Hey end m by hue-ems? C] C] C]

 

VII. 1. wme-mmwemm
Deounaooeoochooe Faun. mm

)

00000003

a fleecing e: Engflsn U
a warm
8. mm or men
(I. W
Other eeeeeenlc sub e.
lens-let emcee: his»
to". science. 100000 I.

WW-
0-

 

 

00000
[3000000
(3000000;

 

 

2. Ieyeudnldlneeneeleleiees?
C) No D Yes-mum?

 

l. "semenlleevenepeeteee'eee?

C] No D Yes-creoeendreeeon

 

4. MmMMdemMeuI-ehm 5. H“ ’0“! child "01‘ bun truud by 8
sentel health professional?

D No C] Yes-MW No Yes - please describe

When one theeenrefleme sled?

Neeetneeenteblemeeneed?

D No D Yes—when?

 

 

157

Child Behavior Checklist (Part 2)
Ages 4 - l6

 

Below is a list of items that describe children. For
each item, think about your child's behavior NOW OR IN
THE PAST SIX MONTHS. Use the following scale to show
how the behavior applies to your child:

SOMEWHAT OR OFTEN TRUE OR
NOT TRUE SOMETIMES TRUE VERY TRUE
0 l 2
l. Acts too young for his/her age
2. Allergy (describe)
3. Argues a lot
4. Asthma
5. Behaves like opposite sex
6. Bowel movement outside toilet
7. Bragging, boasting
8. Can't concentrate, can't pay attention
9. Can't get his/her mind off certain thoughts:
obsessions (describe):
10. Can't sit still, restless, hyperactive
ll. Clings to adults or too dependent
12. Complains of loneliness
13. Confused or seems to be in a fog
14. Cries a lot
15. Cruel to animals
16. Cruelty, bullying, or meanness to others
17. Daydreams or gets lost in his/her thoughts
18. Deliberately harms self or attempts suicide
l9. Demands a lot of attention
20. Destroys his/her own things
21. Destroys things belonging to his/her family or other
children
22. Disobedient at home
23. Disobedient at school
24. Doesn't eat well
.25. Doesn't get along with other children
.26. Doesn't seem to feel guilty after misbehaving
27. Easily jealous
.28. Eats or drinks things that are not food
(describe):
29. Esars certain animals, situations or places other
than school (describe):
30. Fears going to school
.31. Fears he/she might think or do something bad
32. Feels he/she has to be perfect
33. Feels or complains that no one loves him/her
34. Feels others are out to get him/her

35.

Feels worthless or inferior

36.
37.
38.
39.
40.

41.
42.
43.
44.
45.
46.

47.
48.
49.
50.
51.
52.
53.
54.
55.
56.

57.
58.

59.
60.
61.
62.
63.
64.
65.
66.

67.
68.
69.
70.

71.
72.
73.
74.

158

Gets hurt a lot, accident-prone

Gets in many fights

Gets teased a lot

Hangs around with children who get in trouble
Hears things that aren't there
(describe):

Impulsive or acts without thinking
Likes to be alone

Lying or cheating

Bites fingernails

Nervous, highstrung, or tense
Nervous movements or twitching
(describe):

Nightmares

Not liked by other children
Constipated, doesn't move bowels

Too fearful or anxious

Feels dizzy

Feels too guilty

Overeating

Overtired

Overweight

Physical problems without known medical cause:
a. Aches or pains

b. Headaches

c. Nausea, feels sick

d. Problems with eyes (describe):

e. Rashes or skin problems

f. Stomachaches or cramps

g. Vomiting, throwing up

h. Other (describe):

Physically attacks people

Picks nose, skin, or other parts of body
(describe):

Plays with own sex parts in public
Plays with own sex parts too much
Poor school work

Poorly coordinated or clumsy

Prefers playing with older children
Prefers playing with younger children
Refuses to talk

Repeats certain acts over and over: compulsions
(describe):

Runs away from home

Screams a lot

Secretive, keeps things to self

Sees things that aren't there
(describe):

Self-conscious or easily embarrassed
Sets fires

Sexual problems (describe):

Showing off or clowning

75.
76.
77.

78.
79.

80.
81.
82.
83.

84.

159

Shy or timid

Sleeps less than most children
Sleeps more than most children during the day and/or
night (describe):

Smears or plays with bowel movements
Speech problems

(describe):

Stares blankly

Steals at home

Steals outside the home

Stores up things he/she doesn't need
(describe):

Strange behavior

(describe):

85.

86.
87.
88.
89.
90.
91.
92.

93.
94.
95.
96.
97.
98.
99.
100.

101.
102.
103.
104.
105.

106.
107.
108.
109.
110.
111.
112.
113.

Strange ideas
(describe):
Stubborn, sullen, or irritable
Sudden changes in moods or feelings
Sulks a lot
Suspicious
Swearing or obscene language
Talks about killing self
Talks or walks in sleep
(describe):
Talks too much
Teases a lot
Temper tantrums or hot temper
Thinks about sex too much
Threatens people
Thumb-sucking
Too concerned with neatness or cleanliness
Trouble sleeping
(describe):
Truancy, skips school
Underactive, slow moving, or lacks energy
Unhappy, sad, or depressed
Unusually loud
Uses alcohol or drugs
(describe):
Vandalism
Wets self during the day
Wets the bed
Whining
Wishes to be of Opposite sex
Withdrawn, doesn't get involved with others
Worrying

Please write in any problems your child has
that were not listed above:

a.

b.

Ce

160

Child Behavior Checklist (Form II)
Ages 2 - 3

Below is a list of items that describe children. For
each item that describes the child now or within the
last two months, use the following scale to show how the
behavior applies to your child:

SOMEWHAT OR VERY TRUE OR
NOT TRUE SOMETIMES TRUE OFTEN TRUE
0 1 2

‘DQVO‘U'IbUNI-J
e

20.
21.
22.
23.
24.

25.
26.

27.
28.
29.
30.
31.

32.

33.
34.

Aches or pains (without medical cause)

Acts too young for age

Afraid to try new things

Avoids looking others in the eye

Can't concentrate: can't pay attention

Can't sit still or restless

Can't stand having things out of place

Can't stand waiting: wants everything now
Chews on things that aren't edible

Clings to adults: too dependent

Constantly seeks help

Constipated: doesn't move bowels

Cries a lot

Cruel to animals

Defiant

Demands must be met immediately

Destroys his/her own things

Destroys things belonging to his/her family or other
children

Diarrhea or loose bowels when not sick
Disobedient

Disturbed by any change in routine

Doesn't want to sleep alone

Doesn't answer when people talk to him/her
Doesn't eat well

(describe):

Doesn't get along well with other children
Doesn't know how to have fun, acts like a little
adult

Doesn't seem to feel guilty after misbehaving
Doesn't want to go out of the home

Easily frustrated

Easily jealous

Eats or drinks things that are not food
(describe):

Fears certain animals, situations, or places
(describe):

Feelings are easily hurt

Gets hurt a lot, accident-prone

35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.

47.
48.
49.
50.
51.
52.
53.
54.

55.
56.
57.

58.
59.
60.
61.
62.
63.
64.
65.

66.
67.
68.
69.
70.
71.
72.
73.
74.

75.
76.

77.
78.
79.

161

Gets in many fights

Gets into everything

Gets too upset when separated from parents
Has trouble getting to sleep

Headaches (without medical cause)

Hits others

Holds his/her breath

Hurts animals or people without meaning to
Looks unhappy without good reason

Angry moods

Nausea, feels sick (without medical cause)
Nervous movements or twitching

(describe):

Nervous, highstrung, or tense

Nightmares

Overeating

Overtired

Overweight

Painful bowel movements

Physically attacks peOple

Picks nose, skin, or other parts of body
(describe):

Plays with own sex parts too much

Poorly coordinated or clumsy

Problems with eyes (without medical cause)
(describe):

Punishment doesn't change his/her behavior
Quickly shifts from one activity to another
Rashes or skin problems (without medical cause)
Refuses to eat

Refuses to play active games

Repeatedly rocks head or body

Resists going to bed at night

Resists toilet training

(describe):

Screams a lot

Seems unresponsive to affection
Self-conscious or easily embarassed
Selfish or won't share

Shows little affection toward peOple

Shows little interest in things around him/her
Shows too little fear of getting hurt

Shy or timid

Sleeps less than most children during the day
and/or night (describe):

Smears or plays with bowel movements
Speech problems

(describe):

Stares into space or seems preoccupied
Stomachaches or cramps (without medical cause)
Stores up things he/she doesn't need
(describe):

80.

81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.

93.
94.
95.
96.
97.
98.
99.

162

Strange behavior

(describe):

Stubborn, sullen, or irritable

Sudden changes in moods or feelings

Sulks a lot

Talks or cries out in sleep

Temper tantrums or hot temper

Too concerned with neatness or cleanliness
Too fearful or anxious

Uncooperative

Underactive, slow moving, or lacks energy
Unhappy, sad, or depressed

Unusually loud

Upset by new people or situations
(describe):

Vomiting, throwing up (without medical cause)
Wakes up often at night

Wanders away from home

wants a lot of attention

Whining

Withdrawn, doesn't get involved with others
WOrrying

100. Please write in any problems your child has that

were not listed above

APPENDIX K

OBSERVER CHECKLIST

163

Interviewer Observation Checklist

Part A

Interviewer:

Length of Interview:

Patient's attitude toward the interviewer:
Highly cooperative

Cooperative

Indifferent
Uncooperative

bUNH
ee e

Quality of patient interaction:

1. Very responsive
2. Responsive

3. Indifferent

4. Unresponsive

5

. Very unresponsive
Confidence in the validity of the interview:
1. Very confident

2. Moderately confident
3. Not at all confident

164

Part B

Check all the items below that apply to patient's
appearance or behavior during the interview.

5...
see e

\Omfl 01 U1 Ib-wN
ee 0 e e

H
O
O

11.
12.

13.

14.

15.

16.

17.

18.

19.

NOT TRUE SOMEWHAT TRUE VERY TRUE
0 1 2

Face is dirty or unshaven or makeup is smeared
despite availability of washing and shaving
facilities.

Hair is unkempt, tangled, or matted.

Clothes are dirty or in disarray.

Clothes or jewelry or other accessories are bizarre
or grossly inappropriate.

From time to time becomes preoccupied or shows lapse
of attention.

Pays no attention to interviewer a good deal off the
time.

Speech is full of long pauses.

Makes up new words (neologisms).

Abruptly stOps talking in midsentence from time to
time (blocking).

Repeats own words or phrases over and over in a
mechanical manner (e.g. "for the living - for the
living) (perseveration).

Speech is disorganized or incoherent.

Speaks in a faint voice or voice becomes weak and
fades away.

Pitch of voice shows no variation (i.e. completely
monotonous).

Abruptly and rapidly keeps changing the topic of
conversation so that ideas are not completed (flight
of ideas).

Says things in juxtaposition which have little or no
logical or inherent relationship to each other (e.g.
”I'm tired. All peOple have eyes.') (loose
associations).

Combines unrelated words or phrases because they
share similar sounds (e.g. ”I'm sad, bad, mad.')
(clang association).

Keeps deliberately evading answering questions by
becoming inaudible or unintelligible or by changing
the topic or claiming ignorance.

Talks in an ambiguous, obscure, vague, or cryptic
manner.

Talks in a digressive or aimless fashion (rambling).

f

20.

21.

22.
23.
24.
25.
26.

27.
28.
29.
30.
31.

32.

33.

34.
35.

36.

37.
38.
39.
40.
41.
42.

43.
44.
45.
46.
47.
48.

49.
50.

51.
52.

53.
54.

165

Content of answer at times as little if anything to
do with the question asked (irrelevance). (Do not
include wandering or rambling from the topic or
incoherence.)

Keeps adding excessive or useless details to her
answers (circumstantial).

Talks or mutters to herself.

Shouts, yells, or screams.

Keeps cursing or keeps using obscene language.
Makes no verbal response to most or all questions.
Answers questions with single words or brief phrases
only.

Talks on-and on and keeps resisting interruption.
Says only a few words or does not talk at all.
Speaks extremely rapidly and with infrequent pauses.
Speaks extremely slowly.

Has a sad expression or holds her body in a dejected
or despondent posture.

Talks of her condition with practically no outward
sign of emotion.

Laughs or grins inappropriately during discussion of
a serious topic (do not include simple
embarrassment.)

Cries or is tearful.

Maintains a facial expression lacking signs of
emotion.

Shows fleeting and rapidly alternating facial
expressions.

Keeps laughing or giggling in a foolish way.

Often acts inappropriately gay and cheerful.

Has a frightened or apprehensive expression.

Keeps expressing hatred or contempt.

Keeps looking angry.

Keeps virtually the same posture throughout the
interview.

Is slow in all movements.

Keeps scratching, licking, or picking at skin, etc.
Bangs fist on table or stamps feet.

Deliberately tears, throws, or breaks something.
Keeps eyes closed or averted.

Keeps drumming on table with fingers or taps on
floor with foot, etc.

Gets up and moves about restlessly.

Undresses or exposes genitals, or makes an overt
sexual suggestion or advance.

Keeps fidgeting or squirming in her seat.

Assumes strange poses for no apparent reason, or
writhes or contorts her body.

Tries to start an agrument.

Asks or tries to leave room before termination of
interview.

 

55.
56.

57.

58.
59.

60.
61.

62.

63.

64.

65.
66.

67.

166

Makes menacing gesture or physical attack.
Refuses to give further details in a problem are

under discussion.

Expresses objection to interview or resentment at
having to answer questions.

Accuses interviewer of

evil or malicious intent.

Is sarcastic, contemptuous or insulting toward

interviewer.

Resists or refuses to do what is asked of her (e.g.

to sit down.)

Keeps expressing regret for something she has done

or failed to do.

Indicates that because of her behavior, her family,
associates, or authorities are concerned that she
might physically injure herself.
Claims that she is unable to perform at job, school,
or housework or get anything done.

Speaks of contact, power,

knowledge, or sensational

plan which, though not impossible, is extremely
unlikely (non-delusional grandiosity).

Keeps boasting of her accomplishments, skills,
influence, or contacts.
Keeps complaining about herf physical condition or

bodily symptoms.

Manner of response or failure to respond makes it
uncertain whether or not many items are true or

false (e.g. subject,
suggestible).

incoherent,

evasive,

167

Part C

Below are words which describe different kinds of moods
and feelings. Mark an X beside the words which describe
how you think the subject was feeling during the
interview. Check all the words that describe the
subject's feelings. Work rapidly.

l. afraid 22. healthy
2. agreeable 23. hopeless
3. alive 24. irritated
4. angry 25. kindly

5. awful 26. lonely

6. bitter 27. lost

7. blue 28. low

8. calm 29. mad

9. cheerful 30. nervous
10. contented 31. peaceful
ll. COOperative 32. pleasant
12. cruel 33. polite
l3. desperate 34. safe

14. discouraged 35. secure
15. disgusted 36. steady
16. fine 37. strong
17. friendly 38. tense

18. furious 39. terrified
19. good 40. understanding
20. good-natured 41. upset

21. happy 42. worried

 

APPENDIX L

CONSENT FORMS

168

Description of Study

I would like to tell you about the study we are
conducting in the hospital. We are interested in
learning more about mothers and their children. We know
that it is often very difficult for a mother to be away
from her children while she is in the hospital. We also
feel that some mothers may need special help when they
are out of the hospital in taking care of their children
and raising them. In the future, we would like to be
able to deve10p programs that would help mothers with
special needs. However, before we can do that, we need
to know more about mothers, their lives, and problems
they might have in raising their children. Thus, we
want to talk to mothers in the hospital to find out more
about mothers' lives and what might help them.

We are going to be interviewing a number of women
in the hospital and asking them a variety of questions.
The kinds of questions we will be asking mothers include
some questions concerning their own childhood, their
feelings about themselves, their current living
situation, their social relationships, problem behaviors
in their children, and attitudes toward childrearing.

If you decide that you want to participate in this
study, you will be answering these kinds of questions in
an interview that will take from one and a half to two
hours of time. There are some other things about your
participation in the study that I want to tell you about
which are in the consent form. You can ask me any
questions about the study as we go along.

Here's who you can contact if you have
any questions about the study:

Fred Rogosch

Department of Psychology
Michigan State University
E. Lansing, Michigan 48824

Telephone: (517)355-9561

169

CONSENT AGREEMENT

This research study is being conducted by the Psychology
Department of Michigan State University. The study is
concerned with examining ways in which women
hospitalized in inpatient psychiatric facilities deal
with their roles as mothers. We will be asking mothers
a variety of questions concerning their experiences of
their own childhood, their feelings about themselves,
their current living situation, their social
relationships, problem behaviors in their children, and
their attitudes about childrearing. The interview will
take from one and a half to two hours of time.

All information that we collect will be kept strictly
confidential and anonymous. Your name will not appear
on the questionnaires and only group patterns will be
reported rather than information about you as an
individual. You do not have to participate in this
study. If you do agree to participate, you are asked to
sign this form stating that you agree to the following:

1. I understand the explanation of the topic of this
study and what my participation will involve.

2. I understand that my participation in this study is
voluntary. I do not have to participate in this
study if I do not want to. It's up to me to decide.
Whether I participate or do not participate in the
study will not influence the treatment I am
receiving in the hospital. Hospital staff members
will not be informed of my answers.

3. I understand that because my participation in the
study is voluntary, I may skip any question that I
do not want to answer and that I may stOp
participating in the interview at any time without
penalty.

4. I understand that the results of my participation
will be strictly confidential and that my answers
will remain anonymous. No one will be able to know
my answers. Within these restrictions, a summary of
the results of this study about mothers in general
will be given to me when available and if I request
it.

5. I understand that my participation in this study and
the information I give will not be used to alter or

170

influence the relationship that I have with my
children.

6. I understand that there may be no direct benefits to
me as a result of my participation, but that other
people may benefit in the long run because of the
information which is gathered. Further, there are
no anticipated risks to me as a result of my
participation.

7. I understand that the interviewer will also be
reviewing my hospital records, and I have given
permission for this to be done.

8. I understand that I have had an opportunity to ask
any questions about the research study and have them
answered. If I have additional questions about the
study, I may contact Fred Rogosch, Department of
Psychology, Michigan State University, E. Lansing,
MI 48824. Telephone: (517)355-9561.

9. I AGREE THAT I HAVE BEEN GIVEN THE CHANCE TO TALK
ABOUT THE RESEARCH STUDY AND TO ASK QUESTIONS, AND
HEREBY CONSENT TO PARTICIPATE IN THE PROJECT AS
DESCRIBED. I UNDERSTAND THAT I AM FREE TO WITHDRAW
AT ANY TIME.

 

 

PEtientrs Signature Date

 

Witness

 

Witness

171

CONSENT AGREEMENT
Release of Information about Child

1. I understand that additional information is being
requested on my child. I understand that my
decision to allow for provision of this additional
information is voluntary. I can participate in the
interview whether or not I give permission to
release this information about my child.

2. I understand that the person taking care of my child
will provide this information. This person will
know that I have consented to the release of this
information on my child. My participation in this
study will not be discussed because my participation
is confidential.

3. I understand that the answers provided by the person
caring for my child will also be kept strictly
confidential and anonymous with only group results
rather than individual results being reported.

4. I understand that allowing for the provision of this
information about my child will not result in any
direct benefits to me or my child, but that other
people may benefit in the long run because of the
information which is gathered.

5. I understand that a letter will be sent to the
person caring for my child, and that this person
will then be contacted later by telephone to
complete a questionnaire about my child's behavior.

6. I HEREBY GIVE MY CONSENT TO HAVE THE PERSON CARING
FOR MY CHILD CONTACTED TO ANSWER A QUESTIONNAIRE
ABOUT MY CHILD'S BEHAVIOR.

 

 

Signature Date

 

Name of person caring for my child

 

Street

 

City, State ZipCode

( )
Telephone Number

 

 

Witness

 

172

Consent Agreement

This research is being conducted by the Psychology
Department of Michigan State University. We are
interested in learning more about children whose mothers
are in psychiatric hospitals. It is often difficult for
these children to be away from their mothers, and we are
interested in the behavior in general and other possible
problems these children might be having. We are asking
the person taking care of the children of the
hospitalized mother to complete a questionnaire about
the behavior of one of the mother's children.

All information we collect will be kept strictly
confidential and anonymous. Only summaries of group
patterns rather than information about individual
children will be reported. You do not have to
participate in this study. If you do agree to
participate, you are asked to sign this form stating
that you agree to the following:

1. I understand the explanation of the topic of this
study and what my participation will involve.

2. I understand that my participation in this study is
voluntary. My participation will not influence the
treatment the children's mother is receiving in the
hospital. Hospital staff members will not know the
answers I provide.

3. I understand that because my participation in the
study is voluntary, I may skip any question that I
do not want to answer and I may stop participating
at any time without penalty.

4. I understand that the results of my participation
will be strictly confidential and that my answers
will remain anonymous. This also means that the
children's mother will not be able to know the
answers I provide. Within these restrictions, a
summary of the results of this study will be given
to me when available and if I request it.

5. I understand that the answers I provide will not be
used to influence or alter the relationship of the
mother and her children.

6. I understand that there may be no direct benefits to
me as a result of my participation, but that other
peOple may benefit in the long run because of the
information which is gathered. Further, there are

173

no anticipated risks to me as a result of my
participation.

7. I understand that I have an opportunity to ask any
questions about the research study and have them
answered. If I have additional questions about the
study, I may contact Fred Rogosch, Department of
Psychology, Michigan State University, E. Lansing,
Michigan 48824. Telephone: (517)355-9561.

8. I HEREBY CONSENT TO PARTICIPATE IN THE RESEARCH
PROJECT AS DESCRIBED. I UNDERSTAND THAT I AM FREE
TO WITHDRAW AT ANY TIME.

 

 

Signature Date

L IST OF REFERENCES

LIST OF REFERENCES

Achenbach, T. M. (1986). The Child Behavior Checklist
for Ages 2-3. Burlington, VT: Univeristy of Vermont.

Achenbach, T. M. and Edelbrock, C. S. (1983). Manual
for the Child Behavior Checklist and Revised Child
Behavior Profile. USA: Queen City Printers.

Baldwin, A. Le] C0191 R. Bo] & Baldwin, Ce Po (Edge)
(1982). Parental pathology, family interaction, and
the competence of the child in school. Monographs
of the Society for Research in Child Development,
47(5, SeEial No. 197).

 

Baumrind, D. (1967). Child care practices anteceding
three patterns of preschool behavior. Genetic
Psychology Monographs, 22, 43-88.

Baumrind, D. (1971). Current patterns of parental
authority. Developmental Psychology Monograph, 4(1,
Pt. 2).

Beardslee, W. R., Bemporad, J., Keller, M. B., &
Klerman, G. L. (1983). Children of parents with
major affective disorder: A review. American
Journal of Psychiatry, 140, 825-831.

 

Bell, R. Q. (1968). A reinterpretation of the direction
of effects in studies of socialization.
Psychological Review, 22! 81-95.

Bell, R. Q. & Harper, L. V. (Eds.).(l977). Child
effects on adults. Hillsdale, NJ: Erlbaum.

 

Belle, D. (1982). Social ties and social support. In D.
Belle (Ed.), Lives in stress (pp. 133-144). Beverly
Hills: Sage.

174

175

Belle, D. (1984). The impact of poverty on social
networks and supports. In L. Lein & M. B. Sussman
(Eds.). The ties that bind: Men's and womenyg
social networks. Marriage and Family Review, 5(4),
89-103.

Belsky, J. (1980). Child maltreatment: An ecological
view. American Psychologist, 35, 320-335.

 

Belsky, J. (1984). The determinants of parenting: A
process model. Child Development, 55, 83-96.

Belsky, J., Robins, E., & Gamble, W. (1984). The
determinants of parental competence: Toward a
contextual theory. In M. Lewis (Ed.), Beyond the
dyad (pp. 251-279). New York: Plenum Press.

 

Benedek, T. (1959). Parenthood as a develOpmental phase:
A contribution to libido theory. Journal of the
American Psychoanalytic Association, 2, 389-417.

Bogat, G. A., Chin, R., Sabbath, W., & Schwartz, C.
(1983). Social Support Questionnaire. Unpublished
instrument, Michigan State University, Psychology
Department.

Bowen, M. (1978). Family therapy in clinical practice.
New York: Jason Aronson.

 

Bowlby, J. (1969). Attachment and loss. (vol. 1).
Attachment. New York: Jason Aaronson.

Bowlby, J. (1973). Attachment and loss. (vol. 2).
Separation: Anxiety and anger. New York: Jason
Aaronson.

Bowlby, J. (1980). Attachment and loss. (vol. 3). Loss:
Sadness and depression. New York: Jason Aaronson.

 

Boszormenyi-Nagy, I. & Spark, G. M. (1973). Invisible
Loyalties: Reciprocity in intergenerational family
therapy. New York: Bruner-Mazel.

Brown, G. W. (1982). Early loss and depression. In C. M.
Parkes and J. Stevenson-Hinde (Eds.), The place of
attachment in human behavior (pp. 232-268). New
York: Basic Books.

Brown, G. W., Bhrolchain, M. N., & Harris, T. (1975).
Social class and psychiatric disturbance among women
in an urban population. Sociology, 9, 225-254.

 

176

Brown, G. W., Birley, J. L., & Wing, J. K. (1972).
Influence of family life on the course of
schizOphrenic disorders. British Journal of
Psychiatry, 121, 241-258.

Brown, G. W. & Harris, T. (1978). The social origins of
depression. New York: Free Press.

Brown, G. W., Harris, T., & Bifulco, A. (1986). Long
term effects of early loss of parent. In M. Rutter,
C. E. Izard, & P. B. Read (Eds.), Depression in
young people (pp. 251-296). New York: Guilford
Press.

 

Cochran, M. M. & Brassard, J. A. (1979). Child

develOpment and personal social networks. Child
Development, 59, 601-616.

Cohen, J. (1968). Weighted kappa: Nominal scale
agreement with provision for scaled disagreement or
partial credit. Psychological Bulletin, 19, 213-220.

Cohler, B. J. (1984). Parenthood, psychopathology, and
child care. In R. S. Cohen, B. J. Cohler, & S. H.
Weissman (Eds.), Parenthood: A psychodynamic
perspective (pp. 119-147). New York: Guilford Press.

Cohler, B. J., Gallant, D. H., Grunebaum, H. U., Weiss,
J. L., & Gamer, E. (1980). Child-care attitudes and
development of young children of mentally-ill and
well mothers. Psychological Reports, 46, 31-46.

Cohler, B. J., Grunebaum, H. U., Weiss, J. L., Hartman,
C. R., & Gallant, D. H. (1976). Child care attitudes
and adaption to the maternal role among mentally ill

and well mothers. American Journal of
Orthopsychiatry, 46, 123-134.

Cohler, B. J. & Musick, J. S. (1983). Psychopathology of
parenthood: Implications for mental health of
children. Infant Mental Health Journal, 4, 140-164.

Crook, T., Raskin, A., & Eliot, J. (1981). Parent-child
relationships and adult depression. Child
Development, 52, 950-957.

Crockenberg, S. B. (1981). Infant irritability, mother
responsiveness, and social support influences on the
security of mother-infant attachment. Child
Development, 52, 857-865.

177

Cutrona, C. E. & Troutman, B. R. (1986). Social support,
infant temperament, and parenting self-efficacy: A
mediational model of postpartum depression. Child
Development, 5], 1507-1518.

 

Dell, P. F. & Appelbaum, A. S. (1977). Trigenerational
enmeshment: Unresolved ties of single parents to
family of origin. American Journal of
Orthopsychiatry, 47, 52-59.

 

Derogatis, L. R. (1975). Brief Symptoy Inventory.
Baltimore: Clinical Psychometric Research.

 

Derogatis, L. R., Lipman, R., & Covi, L. (1973). The
SCL-90: An outpatient rating scale (Preliminary
Report). PsychopharmacolgyBulletin, 9, 13-28.

 

Fishman, D. B. (1981). A cost-effectiveness methodology
for community mental health centers: Development and
pilot test. Series FN No. 3, DHHS Publication No.
(ADM) 84-767.

Fraiberg, S., Adelson, E., & Shapiro, V. (1975). Ghosts
in the nursery: A psychoanalytic approach to the
problems of impaired infant-mother relationships.
Journal of the American Academy of Child Psychiatry,
_]..'.3’ 387-34le

Frommer, E. A. & O'Shea, G. (1973a). Antenatal
identification of women liable to have problems in
managing their infants. British Journal of
Psyhiatry, 122, 149-156.

 

Frommer, E. A. & O'Shea, G. (1973b). The importance of
childhood experience in relation to problems of
marriage and family building. Bgitish Journal of
Psyhiatry, 123, 157-160.

Garmezy, N. (1974). Children at risk: The search for
antecedents of schizOphrenia. Part I: Conceptual
models and research methods. Schizophrenia Bulletin,
,8, 14-90.

Ghiselli, E. E., Campbell, J. P., & Zedeck, S. (1981).
Measurement theory for the behavioral sciences. San
Francisco: W. H. Freeman.

 

Grunebaum, H., Weiss, J. L., Cohler, B. J., Hartman, C.
R., & Gallant, D. H. (1982). Mentally ill mothers
and their children. (2nd ed.). Chicago: University
of Chicago Press.

178

Hammer, Mel GUtWirthp Lo: & Phillips, Se Le (1982).

Parenthood and social networks. Social Science and
Medicine, £5, 2091-2100.

Harlow, H. F., and Harlow, M. K. (1969). Effects of
various infant-mother relationships on rhesus monkey
behaviors. In B. M. Foss (Ed.), Determinants of
infapg behavior (vol. 4). London: Metheun.

Heard, D. H. & Lake, B. (1986). The attachment dynamic
in adult life. British Journal of Psychiatry, 149,
430-438.

Hereford, C. F. (1963). Changing parental attitudes
through group discussion. Austin: University of
Texas Press.

 

 

Hilgard, J. R. & Fisk, F. (1960). Disruption of adult
ego identity as related to childhood loss of a
mother through hospitalization for psychosis.
Journal of Nervous and Mental Disease, 131, 47-57.

Hoffman, N. G. & Overall, P. B. (1978). Factor structure
of the SCL-90 in a psychiatric pOpulation. Journal
of gpnsulting and Clinical Psychology, 46,
1187-1191. “

Hollingshead, A. B. (1975). Four factor index of social
status. New Haven, Ct.: Yale University, Department
of Sociology.

Kauffman, C., Grunebaum, H., Cohler, B. J., & Gamer, E.
(1979). Superkids: Competent children of psychotic
mothers. American Journal of Psychiatry, 136,
1398-1402.

Klehr, K. E., Cohler, B. J., & Musick, J. S. (1983).
Character and behavior in the mentally ill and well
mother. Infant Mental Health Journal, 4, 250-271.

Lamb, M. E. (1981). The role of the father in child
develOpment (2nd ed.). New York: Wiley.

Leavy, R. L. (1983). Social support and psychological
disorder: A review. Journal of Community Psychology,
E, 3-21e

Leff, J. & Vaughn, C. (1985). Expressed emotion in
families: Its significance for mental health. New
York: Guilford Press.

179

Levitt, M. J., Weber, R. A., & Clark, M. C. (1986).
Social network relationships as sources of maternal
support and well-being. Developmental Psychology,
23, 310-316.

Maccoby, E. E. & Martin, J. A. (1983). Socialization in
the context of the famliy: Parent-child interaction.
In P. H. Mussen (Ed.), Handbook of child psychology
(vol. 4), E. M. Hetherington (Vol. Ed.),
Socialization, personality, and social development
(pp. 1-101). New York: Wiley.

Main, M. & Goldwyn, R. (1984). Predicting rejection of
her infant from mother's representaton of her own
experience: Implications for the abused— abusing
intergenerational cycle. Child Abuse and Neglect, 4,
203-217.

 

Main, M., Kaplan, N., & Cassidy, J. (1985). Security in
infancy, childhood, and adulthood: A move to the
level of representation. Monographs for the Sociepy
for Research in Child DevelOpment, 59(1-2, Serial
No. 209), 66-104.

McLanahan, S. S., Wedemeyer, N. V., & Adelberg, T.
(1981). Network structure, social support, and
psychological well-being in the single parent
family. Journal of Marriage and the Family, 43,
601-612.

 

Mishler, E. G. & Waxler, N. E. (1983). Family processes
and schizophrenia. New York: Jason Aronson.

Morris, D. (1982). Attachment and intimacy. In M. Fisher
8 G. Stricker (Eds.), Intimacy (pp. 305-323). New
York: Plenum Press.

Nguyen, T. D., Attkisson, C. C., & Stegner, B. L.
(1983). Assessment of a patient satisfaction:
DevelOpment and refinement of a service evaluation
questionnaire. Evaluation and Program Planning,Ԥ,
299-314.

Parens, H. (1975). Parenthood as a developmental phase.
Journal of the American Psychoanalytic Association,
a, 154-165.

Patterson, G. R. (1980). Mothers: The unacknowledged
victims. Monographs of the Society for Research in
Child Development, 45(5, Serial No. 186).

 

180

Patterson, G. R. (1986). Maternal rejection: Determinant
or product of deviant child behavior. In W. W.
Hartup & z. Rubin (Eds.), Relationships and
development (pp. 73-94). Hillsdale, NJ: Lawrence

Erlbaum.

 

Pedhazur, E. J. (1982). Path analysis (Ch. 15).
Multiple regression in behavioral research. (2nd
ed.). New York: Holt, Rinehart, & Winston.

Raskin, A., Boothe, H. H., Reatig, N. A.,
Schulterbrandt, J. G., & Odle, D. (1971). Factor
analyses of normal and depressed patients' memories
of parental behavior. Psychological Reports, £2,
871-879.

Ricks, M. H. (l985). The social transmission of parental
behavior: Attachment across generations. Monographs
for the Society for Research in Child DeVElOpment,
§Q(l-2, Serial No. 209), 211-227.

Rogosch, F. A. & Mowbray, C. T. (1987). The maternal
role of inpatient mothers. Unpublished manuscript,
Michigan Department of Mental Health.

Rosenberg, M. (1965). Society and the adolescent self
image. Princeton: Princeton University Press.

Roth, R. M. (1980). The mother-child relationship
evaluation. Los Angeles: Western Psychological
Services.

 

Rutter, M. (1981). Maternal deprivation reassessed (2nd
ed.). New York: Penguin Books.

Rutter, M. & Madge, N. (1976). Cycles of disadvantage.
London: Heinemann.

Salzinger, 8., Kaplan, S., & Artemyeff, C. (1983).
Mothers' personal social networks and child
maltreatment. Journal of Abnormal Psychology, 92,
68-76. '—

Sameroff, A. & Chandler, M. (1975). Reproductive risk
and the continuum of caretakeing casualty. In F. D.
Horowitz (Ed.), Review of child development
research (vol. 4) (pp. 187-244). Chicago: Univeristy

of Chicago Press.

181

Sameroff, A. J., Seifer, R., & Zax, M. (1982). Early
development of children at risk for emotional
disorder. Monographs of the Society for Research in
Child DevelOpment, 41(7, Serial No. 199).

 

Sarason, B. R., Shearin, E. N., Pierce, G. R., &
Sarason, I. G. (in press). Interrelationships of
social support measures: Theoretical and practical
implications. Journal of Personality and Social
Psychology.

Schaefer, E. S. (1963). Children's reports of parental
behavior: An inventory. Child Development, 36,

 

Schaefer, E. S. & Bayley, N. (1967). Validity and
consistency of mother-infant observations,
adolescent maternal interviews, and adult
retrospective reports of maternal behavior.
Proceedings of the 75th Annual Convention of the
American Psychological Association, 1, 147-148
(SummaryTTi

Stollak, G. E., Scholom, A., Kallman, J., & Saturansky,
C. (1973). Insensitivity to children: Responses of
undergraduates to children in problem situations.
Journal of Abnormal Child Psychology, 2(1), 169-180.

Test, M. A., & Berlin, S. B. (198l). Issues of special
concern to chronically mentall ill women.
Professional Psychology, 13, 136-145.

Teyber, E. C., Messe, L. A., & Stollak, G. E. (1977).
Adult responses to child communications. Child
DevelOpment, 42! 1577-1582.

 

Thomas, A., Chess, 8., & Birch, H. G. (1968).
Temperament and behavior disorders in children.
London: University of London Press.

watt] N. F.) Aflthony, E. J., WYDDEI L. C.) & Ralf, J. E.
(Eds.), (1984). Children at risk for schizophrenia.
New York: Cambridge University Press.

Wahler, R. G. (1980). The insular mother: Her problems
in parent-child treatment. Journal of Applied
Behavior Analysis, 13, 207-219.

182

Wahler, R. G., Leske, G., & Rogers, E. S. (1979). The
insular family: A deviance support system for
oppositional children. In L. A. Hamerlynck (Ed.)
Behavioral systems for the developmentally disabled:
I. School and family environments (pp. 102-127). New
York: Brunner/Mazel.

Weissman, M. M., Paykel, E. S., & Klerman, G. L. (1972).
The depressed woman as mother. Social Psychiatry, 1,
98-108.

Wolkind, S., Hall, F. & Pawlby, S. (1977). Individual
differences in mothering behavior: A combined
epidemiological and observational approach. In P. J.
Graham (Ed.), Epidemiological approaches in child
psychiatry (pp. 107-123). New York: AcademiciPress.

 

Wynne, L. C. (1981). Current concepts about
schizophrenia and family relationships. Journal of
Nervous and Mental Disease, 169, 82-89.

"‘mylumummES